MPAs typically seek to protect the ocean from various extractive activities, such as fishing and mining, and have shown to provide conservation benefits (Lester & Halpern, 2008), particularly when effectively managed (Gill et al., 2017). But they provide little protection against increasing global-scale threats like climate change – which can only be mitigated by a major global reduction in greenhouse gas emissions (Bates et al., 2019).

All the same, considering climate change impacts in the design of MPAs can give them the best chance at protecting marine biodiversity and ecosystem functions into the future. Incorporating climate change into MPA design is a significant challenge, but numerous guidelines are available (see Related tools and resources, Jones et al., 2016, Fredston Hermann et al., 2018, and Tittensor et al., 2019).

Here, we provide an overview of MPA design approaches that may result in more effective MPAs in the face of climate change. We say “may” as there is a great deal of uncertainty about the benefit of MPAs in improving the resilience of marine ecosystems to climate change (Mcleod et al., 2019) and the impact of “climate-smart” design approaches (Micheli et al., 2012; Selig et al., 2012) (see Module 4 MPA monitoring in the face of climate change Q&A).

There is certainly no “one size fits all” approach – the best approach will depend on the ecological and socioeconomic context and scale of the region. These approaches are intended to be suggestions rather than prescriptions, and not all suggested approaches can be applied to all MPA design processes.

Is this not for you?

I am a using MPAs to and I need help to by

Climate-smart MPA design

We discuss three general categories of climate-smart protected area design:

Protect future habitat
Protect places predicted to experience relatively low climate impacts
Employ best-practice design principles.

Within each category, we discuss a selection of approaches that have been published in the peer-reviewed literature and, in some cases, applied to help design real-world MPAs. Importantly, while we use these three general categories for simplicity, they are not mutually exclusive, and can be combined in many ways to improve MPA resilience.

Protect future habitat

MPAs today are often “fixed” or “static”. As the climate changes, habitat and species distributions, condition and/or composition within MPA borders are also likely to change, which may make static MPAs ineffective. Two ways to try to account for these changes are to:

Establish MPAs today where habitats and species are likely to shift in the future (Arafeh‐Dalmau et al., 2021)
Plan for dynamic MPAs to move as species/habitats move (Game et al., 2009; Tittensor et al., 2019).

Predicting where and when habitats and species are likely to move can be data intensive (especially when multiple species are involved), technically challenging and fraught with error. Species are likely to shift at different rates and in different directions, which can make protecting diverse species assemblages – often a goal of MPAs – difficult.

Ideally, species distribution or niche/bioclimatic models would be used to predict future species distributions, which would then be incorporated into a decision-support tool such as Marxan. However, if data and/or technical capacity are limited, Fredston‐Hermann et al. (2018) provide some general rules to consider as outlined in Box 1.

MPA distribution general rules

Large MPAs should be designed to be “tall” rather than “wide” (spanning north to south instead of east to west) to accommodate marine species generally moving poleward (Cashion et al., 2020).
For MPAs intended to protect a single species, siting MPAs closer to the leading (often poleward) edge of the geographic range, as compared to the trailing edge, may keep that species within MPA boundaries for longer, effectively buying time as the climate changes (Fredston Hermann et al., 2018).
Multiple MPAs that make up a network larger than most species ranges can act as “stepping stones” (Fredston Hermann et al., 2018). These MPAs do not have to be implemented all at once and can even be additions to existing MPA networks (Alagador et al., 2014).

Establishing MPAs where current and future species ranges intersect or are in close proximity can have a similar effect (e.g., Makino et al., 2014). So too can protecting areas that are expected to aggregate populations in the face of climate change such as migration corridors, poleward habitat edges and thermal refugia (discussed below), and populations with high genetic diversity.

None of this should distract from the urgent need to reduce greenhouse gas emissions and to keep average global temperature rise to below 1.5°C. Stabilizing the climate is the most effective way to reduce the potential for species shifts and related costs.

Planning for MPAs to move as species/habitats move (“dynamic” MPAs) is a relatively simple concept, but currently has few examples in practice. Dynamic MPAs could face legal barriers and be challenging to establish, manage and enforce. Also, species are predicted to move at different rates, some fast and some slow (Poloczanska et al. 2013). Given that some species may see distribution shifts of up to 400km per decade, this may involve coordinating across multiple countries, which is challenging.

Dynamic MPAs could support climate change resilience in several ways:

Rotating protection (Game et al., 2009)
Tracking desirable environmental conditions (e.g., sea surface temperature) or species ranges (Hobday, 2011)
Supplementing permanent, “static” MPAs to create more flexible networks.

Effectively managing and enforcing dynamic MPAs may pose challenges when MPA boundaries shift. These should be considered in estimates of implementation costs (e.g., costs of establishment, management, and opportunity costs to stakeholders) and throughout the design process. For example, shifting MPA boundaries would require further communication to stakeholders and may cause compliance breaches (intentional or unintentional) as users adjust to new boundaries. Existing guidance on considering costs in MPA design and implementation can be adapted for dynamic MPAs (e.g., Ban and Klein, 2009).

If shifting MPAs are implemented in stages, there may be political hurdles as governments and funding opportunities often operate on short, sporadic timescales. Ensuring adequate policy frameworks and funding mechanisms for effective, long-term dynamic MPAs is critical.

Notably, “dynamic” protection is often common in community-based conservation approaches, such as seasonal closures for fisheries management. For example, seasonal no-take closures have been used for sustainable octopus fishing in Madagascar (Humber et al., 2006). Many species are managed with “closed” seasons where targeted catch is prohibited as part of their management strategy. Lessons from these conservation tools could be used to help scale up actions for larger-scale dynamic MPA networks.

Protect places expected to experience relatively low climate impacts

Not all marine areas are or will be impacted equally by climate change. Identifying and protecting areas that have relatively lower climate risk (sometimes called “refugia”) is a strategy that is gaining attention. It may:

Avoid protecting areas with high levels of stressors that MPAs cannot mitigate, such as those associated with climate change (Klein et al., 2013).
Spread and/or reduce risk of extreme damage across MPAs from weather events, such as cyclones (Allison et al., 2003; Game et al., 2008).
Result in greater resilience and regeneration potential if local threats within climate refugia are also effectively managed (Beyer et al., 2018; Hoegh-Guldberg et al., 2018) (see Figure 1 below).

Refugia can be identified using different approaches. For example, climate velocity data is freely available for the entire ocean and can be used to identify slow-moving places, or refugia (e.g. Arafeh‐Dalmau et al., 2021).

Globally, lower climate risk marine areas are generally thought to be at higher latitudes, deeper areas and upwelling regions where warming is often less (Kavousi, 2019). However, recent studies have shown that the deep ocean is also highly exposed to the effects of warming (Brito-Morales et al., 2020). Further consideration of other (or multiple) climate impacts, like ocean acidification, is also needed to identify less exposed areas (Kavousi, 2019; Tittensor et al., 2019).

Because low climate impact places can be defined and identified in many ways, it’s important to incorporate other “climate-smart” design principles to build an MPA network with a better chance to withstand future change (Mcleod et al., 2009; Beyer et al., 2018). These include spreading the risk by protecting a greater diversity and/or number of sites, and incorporating connectivity (Mcleod et al., 2009).

Figure 1. Schematic of conservation action for protecting less climate exposed reefs. The diagram emphasises that conservation action will be futile without strong global action on climate change (A). The diagram then describes the processes for identifying less climate exposed reefs (B - prioritize investments), identifying core threats and solutions in these areas to target conservation action, and implementing action in collaboration with local stakeholders to reduce and manage local-scale threats. Source: Hoegh-Guldberg et al., 2018. CC BY-NC-ND 4.0.

Employ best-practice design principles

Numerous design principles have been developed to help MPAs achieve a broad range of goals related to biodiversity, fisheries and climate change (Mcleod et al., 2009; Green et al., 2013). Mcleod et al., 2009 and Mcleod et al., 2019 summarize recommendations for designing MPA networks that address the impacts of climate change, including size, shape, risk spreading, critical areas, connectivity, ecosystem function and ecosystem management.

Importantly, some of these principles, such as protecting a diverse range of habitats and replicating habitats (representation), are simple and can easily be applied to data-limited regions at a range of scales. Other approaches (e.g., connectivity or the movement of energy, materials or organisms that support life on Earth) can be more data and methodologically intensive.

Incorporating connectivity into MPA design in the face of climate change has been explored extensively (see Magris et al., 2014; Carr et al., 2017), but is challenging for many reasons. These include a lack of data and required expertise, the challenge of defining and quantifying the type of connectivity and how this might change through time, and defining parameters to limit the temporal and spatial scales for prioritization. Due to this complexity, many MPA designs revert to general rules, such as those suggested in Mcleod et al., 2009, summarized below (see Table 1 of Mcleod et al., 2009 for more detail):

Size, placement and number of protected areas

General guidelines suggest MPAs should be at least 10–20 km in diameter, and large enough to protect a sample of all marine habitat types and ecological processes.
Shape and/or orientation

Using simple shapes, like squares and rectangles, minimises the number of edges (i.e., edge effects) while maximising the area protected.
Spreading the risk of failure

To do this, MPAs should represent at least 20–30% of each habitat, protect multiple samples of each habitat type (replication), and make sure there is enough space between individual MPAs to spread risk from disturbance events, such as cyclones. All of these principles should also be applied across temperature regimes to reduce risks from climate change.
Protecting critical areas

This includes nursery grounds, spawning aggregations, areas of high species diversity, and areas that will be least impacted or most likely to survive in the face of climate change.
Adding specific linkages between areas (e.g., corridors)

The main biological connectivity patterns to consider are larval dispersal, adult movement, ecosystem connectivity, and future connectivity. Generally, it is recommended to space MPAs no more than 15–20 km apart to ensure larval replenishment, to protect larger areas to capture adult movement, to protect adjacent habitats like seagrass and mangroves, and to protect places that may benefit from connectivity in the future (i.e., places connected to climate refugia).

Maintaining key functional groups, such as herbivores on coral reefs.
Improving management across realms (e.g., “ridge to reef”, ecosystem-based management).

However, scientific recommendations to promote reef resilience in the face of climate change do not come without challenges. Mcleod et al., 2019 lays out recommendations, evidence and challenges of resilience-based management of coral reefs (Table 1). Resilience-based management focuses on the socio-ecological system of coral reefs (i.e., reefs and the people that depend on them) to support stress tolerance, recovery and adaptation.

Advice, evidence and challenges for management that supports reef resilience.

While resilience-based management is broader than MPA implementation, the recommendations here can still be applied to MPAs under climate change.

Protect a diversity of species, habitats, and functional groups Maintain response diversity and redundancy within functional groups

Evidence for resilience principles:

A high diversity of coral habitat types helps to ensure that a wide range of functional traits, due to adaptations to local environmental conditions, are represented within the seascape (Nyström et al., 2008).
Diverse functional groups of herbivores can enhance coral recovery (Bellwood et al., 2006, Mumby et al., 2007, Burkepile and Hay, 2008, Steneck et al., 2014).
Spatial heterogeneity may reduce the risk for catastrophic regime shifts (van Nes and Scheffer, 2005).

Challenges for management:

Managing to support diverse species may, in some cases, increase the proportion of sensitive species
The ability of herbivores to promote recovery is highly context-dependent (McClanahan et al., 2002, Aronson and Precht, 2006) and depends on their functional role, the algae that they graze (Graham et al., 2015), and diversity of species responses to environmental change (response diversity; Elmqvist et al., 2003) and other anthropogenic stressors (Edwards et al., 2014, Adam et al., 2015, Graham et al., 2015).

Reduce reef stressors (e.g., pollution, sedimentation, physical impacts)

Reducing stressors is important to reef resilience because nutrient pollution, sedimentation. Impacts will be context-specific, based on local ecological and oceanographic conditions, and different sensitivities of corals to and physical impacts can cause adverse impacts to corals:

Reduced thermal tolerance of corals (Fabricius, 2005, Wooldridge, 2009, Humanes et al., 2017)
Decreased recovery potential (Connell et al., 1997, Fabricius, 2005, Carilli et al., 2009, Fourney and Figueiredo, 2017, Humanes et al., 2017)
Slower coral growth rates (McManus et al., 1997, Fox et al., 2005, Shantz and Burkepile, 2014)
Increased coral disease and bleaching (Vega Thurber et al., 2014)
Increased algal growth (Jompa and McCook, 2002)

Maintaining water quality can support coral reef resilience (Bellwood et al., 2004, Wooldridge, 2009).

Challenges for management:

Impacts will be context-specific, based on local ecological and oceanographic conditions, and different sensitivities of corals to nutrients, sediments, and physical impacts (Fabricius, 2005, Erftemeijer et al., 2012).
Impacts may not always be negative, e.g., physical impacts can, in some cases, create new habitat for coral settlement (Chabanet et al., 2005).

Implement MPAs to support reef resilience including the protection of refuges (e.g., area less vulnerable to climate impacts)

MPAs are an important tool to support reef resilience as they can help to:

Reduce local stressors (Bellwood et al., 2004, Wooldridge, 2009)
Restore coral reef food webs (Bellwood et al., 2004)
Reduce outbreaks of coral predators (Sweatman, 2008)
Reduce coral loss (Mellin et al., 2016, Wolff et al., 2018), and Support herbivory which can facilitate coral recruitment (Mumby et al., 2007, Selig and Bruno, 2010).
Promote recovery (Mumby and Harborne, 2010, Micheli et al., 2012, Perry et al., 2015)

Challenges for management:

The role of MPAs in protecting coral reefs from climate change is equivocal (Côté and Darling, 2010, Hughes et al., 2017, Roberts et al., 2017), and there is only limited, regional evidence that MPAs are outperforming other areas (Selig and Bruno, 2010).
The number of potential refuges is likely to decline through repeated bleaching events (Hughes et al., 2017).

Maintain pathways of connectivity

Evidence for resilience principles:

Connectivity can promote reef recovery by providing a supply of coral larvae from less impacted locations (Mumby and Hastings, 2008, Jones et al., 2009, Hock et al., 2017).

Challenges for management:

Connectivity may not always support recovery (Graham et al., 2011) and in some cases, it may facilitate the spread of invasive species or pollutants, thus, its role in supporting resilience depends upon the local context (McClanahan et al., 2002).

Manage adaptively to accommodate uncertainty and change

Evidence for resilience principles:

Key components of adaptive management include: monitoring and evaluation; a continuing cycle of experimentation and reevaluation; participatory approaches; and diverse stakeholder participation (Ostrom, 1990, Schreiber et al., 2004, Folke et al., 2005, Biggs et al., 2012).

Challenges for management:

Barriers to adaptive management include: lack of resources for monitoring; unwillingness to embrace uncertainty; lack of data on key processes (e.g., recruitment), and perceived as expensive and/or ecologically risky or as a threat to existing research programs and management regimes (Walters, 2007).

Prioritize areas with low environmental risk and high social adaptive capacity

Evidence for resilience principles:

Managers should prioritize areas with low environmental risk and high social adaptive capacity (McClanahan et al., 2008).

Challenges for management:

Areas of high environmental risk may be important when they also include high adaptive capacity -such areas may drive the development of innovation (e.g., in restoration practices; McClanahan et al., 2008).

Incorporate social and ecological indicators to assess early warnings, recovery patterns, and regime shifts in conservation planning and monitoring

Evidence for resilience principles:

By combining projected future exposure with data on resilience indicators (e.g., McClanahan et al., 2012), managers can map relative vulnerability to climate change and prioritize actions to support reef resilience (van Hooidonk et al., 2016).
Social drivers (e.g., technologies, markets, demographic changes, and changes in governance structures or policy) can provide early indicators of regime shifts (Hicks et al., 2016).

Challenges for management:

Static measures (coral cover, fish abundance, diversity) can be poor indicators of resilience (Mumby et al., 2014a, Mumby et al., 2014b); high coral cover may fail to indicate changes in reduced recruitment potential or reduced herbivory (Anthony et al., 2015).
Factors that determine recovery potential (reef structural complexity, water depth, herbivorous fish biomass, nutrient regime, and coral recruit densities) differ across regions making it challenging to prioritize management actions that support recovery (Graham et al., 2015).

Invest in experimental approaches to support resilience (e.g., enhance the natural adaptive capacity of reef organisms via assisted evolution)

Evidence for resilience principles:

Some corals in areas with wide temperature fluctuations resist stress better than corals from less extreme environments (e.g., transplanted corals into hotter and more variable conditions acquired thermal tolerance demonstrating short-term acclimatization and longer-term adaptive acquisition of climate resistance; Palumbi et al., 2014)
Enhancement of the adaptive capacity of corals to warming through assisted evolution and experimental management trials can help promote the resilience and climate tolerance of key species (Van Oppen et al., 2017).

Challenges for management:

Corals differ in their ability to adapt or acclimatize (Baker et al., 2004), e.g., some can increase their proportion of heat-resistant symbiont types (Clade D providing an increased tolerance of∼ 1–1.5 °C; Berkelmans and van Oppen, 2006), while others cannot.
Large potential for unintended consequences and cannot be implemented at the scales necessary to reverse reef decline

Implement strategies to build social and ecological adaptive capacity (e.g., maintaining diversity of human opportunities and economic options that encourage adaptation and learning; broaden stakeholder participation)

A key aspect of resilience in social-ecological systems is the flexibility of resource users to switch from one livelihood strategy to another (Berkes and Seixas, 2006, Cinner et al., 2009b) as it provides the conditions necessary for experimentation and the ability to respond to change (Olsson et al., 2004; Tompkins and Adger, 2004; Berkes and Seixas, 2006, Marshall et al., 2007).
The participation of diverse stakeholders can improve legitimacy, facilitate monitoring and enforcement, promote understanding of system dynamics, improve a management system's capacity to respond to shocks and disturbances (Ostrom, 1990, Folke et al., 2005, Reed, 2008), and facilitate collective action (Schreiber et al., 2004, Biggs et al., 2012).

Challenges for management:

Livelihood diversification does not necessarily indicate high adaptive capacity – in some cases, it may reflect a low standard of living that requires increased effort to meet basic needs; in other cases, it may be a deliberate strategy to “spread the risk” and improve resilience to fluctuations/shocks (McClanahan and Cinner, 2011).

Implement strategies to facilitate adaptation and transformation (e.g., through promoting polycentric governance systems; providing buffer zones around protected sites)

Polycentric structures may support functional redundancy (e.g., national governance can step in if local governments fail to achieve desired outcome, or vice versa; Rohlf, 1991, Biggs et al., 2012).
Polycentric systems can provide opportunities for learning and experimentation, broader levels of participation in governance, and consideration of local knowledge and knowledge sharing across scales (Olsson et al., 2004, Ostrom, 2005, Biggs et al., 2012).
Buffer zones around protected areas can increase the potential of species to adapt and move.

Challenges for management:

The long-term robustness of large-scale polycentric governance systems has recently been challenged (Morrison, 2017).
Improved data on how species ranges shift in response to climate change is needed to inform location of buffer zones (Smith and Lenhart, 1996).

Summary of climate-smart MPA design approaches

Table 2 summarizes the strengths and limitations of each climate-smart protected area design approach. Importantly, a limitation of all approaches is that we do not have a good understanding of their relative costs (design/implementation) and benefits. For example, we cannot design MPAs with and without each approach and evaluate their performance (e.g., how well they protect marine biodiversity under different climate futures).

Strengths and limitations of the three climate-smart protected area design strategies outlined. Adapted from Jones et al. (2016), which contains an extended list of planning approaches.

Protect future habitat

Strengths:

Applicable to a wide range of species and spatial scales
Sea surface temperature data that can be used to predict shift in species’ range is globally available

Limitations:

Required data may not be available or at a fine enough scale – particularly for rare or small-ranged species
Modelling decisions can greatly influence results
Requires technical expertise
Predictive models based on current species data may be inaccurate, as species respond individually to climate change in the future

Example applications:

We are unaware of any practical examples, but believe the approach merits mention because it has a strong theoretical underpinning. For example, protecting future habitat has been explored in Makino et al. (2014), accommodating future coral range expansions based on projections of future sea surface temperature.

Protect places predicted to experience relatively low impacts

Strengths:

Can be identified from historical and/or current climatic factors
Can include, but does not require, predictions of climate or species distributions
Useful for large-scale prioritizations as global data exists (e.g. cumulative impact maps (Halpern et al. 2019) and climate velocity (Brito-Morales et al. 2020))

Limitations:

Difficult to plan for specific conservation features (e.g., identifying species-specific refugia)
Requires some technical expertise that may require expert consultation
Gaps and uncertainty in data and future impacts may lead to inaccuracies

Example applications:

The Nature Conservancy (TNC) has incorporated climate refugia into MPAs in the US, Pacific, Asia and the Caribbean

WWF’s Coral Reef Rescue Initiative is working to protect coral reefs across seven countries that have been identified to be less exposed to climate impacts and have higher connectivity to nearby reefs (Beyer et al., 2018). While no MPAs have been specifically designated currently, this is a major aim of the initiative.

Arafeh‐Dalmau et al. (2021) demonstrated this approach using climate velocity data in the Mediterranean Sea

Employ best-practice design principles

Strengths:

Many principles are relatively simple and do not rely on complicated data (size, spacing, replication)
Often principles align with other goals laid out in national and international protected area targets (e.g., representation)

Limitations:

Indirectly considers potential aspects of climate change

Example applications:

In Kimbe Bay, Papua New Guinea (Green et al., 2009), and in California, USA, planners applied many similar design principles (e.g., replication of habitats protected in the MPA network (Saarman et al., 2013)).

Limitations and gaps surrounding climate-smart MPA design

While significant advances have been made in recent years to predict and plan for climate change, there are still many gaps in our understanding. There is still much uncertainty around how both conservation actions and climate change impacts act across local, regional, national and multinational scales. The mismatch between the global responsibility and impacts of greenhouse gas emissions and local conservation action can be challenging.

Timeframes over which practitioners should plan and act are often uncertain or not well aligned with the timescales of predictive modelling approaches and/or government cycles. During the design process, stakeholders and those designing MPAs should work together to determine relevant timescales to assess impacts to best address their needs now and into the future.

Ongoing strategic monitoring, adaptive management and financing are essential to ensure MPAs are achieving their intended objectives (e.g., climate refugia are functioning as refugia, etc.). A detailed description of stakeholder engagement in the design of the Tun Mustapha MPA in Malaysia gives practical advice on support and training for MPA practitioners in best practice MPA design (Jumin et al., 2018).

The static nature of MPAs is also a limitation. Policy and legal frameworks for MPAs will likely need to become more flexible in the future to best adapt to a changing climate. This could pose management challenges if borders and objectives shift, as well as increase costs of enforcement and adaptive management. MPA managers could consider incorporating more flexible conservation tools, such as those used in fisheries management and other effective area-based conservation measures (OECMs).

Finally, the human response to climate change needs to be considered in MPA design. This might include:

Potential changes in natural resource use (e.g., food production)

Changes in human demographic and migration patterns

Other responses to climate impacts (e.g., hard engineering solutions).

Modelling human behaviour is difficult and methods and data are often lacking to develop robust socioecological models. At the very least, engaging a representative group of stakeholders and documenting the full suite of potential direct and indirect climate change impacts when designing an MPA in a given area is a good place to start.

How can MPA design account for climate change?

Where in the MPA lifecycle?

What's this about

Good for ...

Is this not for you?

Climate-smart MPA design

Protect future habitat

MPA distribution general rules

Protect places expected to experience relatively low climate impacts

Employ best-practice design principles

Advice, evidence and challenges for management that supports reef resilience.

Summary of climate-smart MPA design approaches

Strengths and limitations of the three climate-smart protected area design strategies outlined. Adapted from Jones et al. (2016), which contains an extended list of planning approaches.

Limitations and gaps surrounding climate-smart MPA design

A partnership between

How can MPA design account for climate change?

Where in the MPA lifecycle?

What is the MPA lifecycle?

What's this about

Good for ...

Is this not for you?

Climate-smart MPA design

Protect future habitat

MPA distribution general rules

Protect places expected to experience relatively low climate impacts

Employ best-practice design principles

Advice, evidence and challenges for management that supports reef resilience.

Summary of climate-smart MPA design approaches

Strengths and limitations of the three climate-smart protected area design strategies outlined. Adapted from Jones et al. (2016), which contains an extended list of planning approaches.

Limitations and gaps surrounding climate-smart MPA design

A partnership between