Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Spatial socioeconomic data as a cost in systematic marine conservation planning

Spatial socioeconomic data as a cost in systematic marine conservation planning Introduction In planning for conservation, planners and scientists have focused on the biological benefits of conservation plans (i.e., Naidoo 2006 ). There is increasing recognition that conservation objectives must be achieved efficiently because of limited conservation resources. Hence, socioeconomic costs must be integrated into conservation planning ( Naidoo 2006 ; Carwardine 2008b ), where “cost” is intended to reflect the socioeconomic impacts of conservation areas. There has been an influx of initiatives to promote the design and implementation of marine protected areas (MPAs) around the globe ( Spalding 2008 ) and many marine planners are seeking guidance on how to use socioeconomic data. By explicitly incorporating socioeconomic costs into systematic marine conservation planning, we can avoid costly conservation mistakes. Systematic conservation planning is an approach that guides the location and design of conservation areas that achieve explicit biodiversity objectives ( Margules & Pressey 2000 ). Application of this approach to real‐world conservation assessments is supported by decision support tools, e.g. , Marxan ( Ball & Possingham 2000 ; Possingham 2000 ), C‐Plan ( Pressey 2009 ), Zonation ( Moilanen 2005 ), and ResNet ( Kelley 2002 ). These tools were designed to support, not make, decisions on the location of conservation areas; they provide the basis for discussions but do not provide an answer that is to be unequivocally accepted ( Possingham 2006 ). A common problem planners aim to solve is to minimize the cost of conservation areas while achieving quantitative representation targets ( Airamé 2003 ; Stewart & Possingham 2005 ; Naidoo 2006 ; Possingham 2006 ; Klein 2008b ). The Marxan software was designed to solve this problem. Recent literature highlights the importance of incorporating socioeconomic costs into conservation planning for two main reasons. First, including socioeconomic costs minimizes the impacts on resource users ( Klein 2008a, 2008b ) and thus reduces conflicts between resource users and conservationists ( Scholz 2004 ; Crawford 2006 ; Carwardine 2008b ; Klein 2008a ). Second, including such costs produces plans that are cost‐effective to implement and manage ( Naidoo 2006 ; Carwardine 2008b ). We summarize the costs of marine conservation, review how these costs have been used in conservation planning, and discuss future research priorities to address shortcomings of current approaches. Reviews on incorporating costs into conservation planning have been conducted for terrestrial planning ( Hughey 2003 ; O’Connor 2003 ; Polasky 2005 ; Naidoo 2006 ). A review highlighting the unique aspects of socioeconomic costs in marine planning is necessary because: (1) aspects of marine systems differ from their terrestrial counterpart (see below); (2) the field has progressed quickly over a short time; and (3) many practical applications of these concepts are difficult to find in the literature yet useful and sought after by marine planners. Marine conservation costs Many costs are associated with conservation planning, including acquisition costs, management costs, transaction costs, and opportunity costs (see Naidoo 2006 ). Acquisition costs are atypical in the marine environment because waters are not usually privately owned. Management costs are associated with enforcing and maintaining MPAs ( Naidoo 2006 ). Transaction costs are associated with negotiating protection, such as the time and staff involved in stakeholder negotiations. Opportunity costs are foregone revenues, e.g., the value to fisheries and other marine uses ( Naidoo 2006 ), and are commonly used to site MPAs. Damage costs are associated with loss to economic activities due to conservation activities; an example from land is elephants or monkeys from reserves raiding agricultural fields ( Naidoo 2006 ). The relative importance of these types of costs differs in terrestrial and marine systems. For example, the location of MPAs is predominantly influenced by the opportunity cost to industries (e.g. , fishing, mining). Although opportunity costs are also important in terrestrial planning, an additional cost is associated with the land tenure (e.g. , acquisition or stewardship costs). Land tenure or property rights systems influence the types of costs that predominate. Marine systems are typically common‐pool resources ( Ostrom 1999 ), whereas terrestrial systems have a much higher proportion of private ownership, although exceptions exist (e.g. , Beck 2004 ). Therefore, acquisition costs equivalent to the purchase of land are rare in the marine environment. Management and transaction costs are likely to be of similar importance, but little information exists on such costs ( Balmford 2004 ). We have not encountered any mention of damage costs due to marine conservation. Critical to identifying costs for MPAs is clearly defining the objective and specifying feasible conservation actions (e.g., no‐take marine reserve, multi‐purpose conservation area) ( Carwardine 2008a ). For example, if the objective is to select MPAs while minimizing the effect of such closures on commercial fishers, then the most suitable opportunity cost would be a measure of their impact on these fisheries. The cost impacts the spatial location of MPAs ( Figure 1 ). Using inappropriate cost measures can lead to costly conservation mistakes ( Carwardine 2008b ). For example, Carwardine (2008b) found that biodiversity targets can be up to twice as expensive to achieve if incorrect cost data are used. The costs identified vis‐à‐vis the objective have to be summarized into a single cost unit in one layer because the current formulation used in decision‐support tools (e.g., Marxan, C‐Plan, Resnet) allows for only one cost layer. 1 Results of scenarios using different costs. The left column shows the importance to conservation of areas under a variety of cost scenarios, and the right column depicts the corresponding costs: (a) area as a cost, (b) human impacts as a cost, and (c) commercial fisheries as a cost. The maps depict scenarios that represent 10% of every feature. Reproduced from Ban (2009b) . For use in planning, cost data have to be spatially explicit at a scale fine enough to differentiate areas. When spatially explicit cost data are not available, surrogates can be used (e.g., global fishing effort, or coastal population density as a proxy of fishing effort, Ban 2009a ). Even when spatially explicit costs are available, these often represent surrogates for the real costs of conservation planning. For example, a marine plan might want to minimize the monetary cost of establishing MPAs, but uses areas of relative importance to commercial fisheries as the surrogate. How well such surrogates represent actual cost efficiencies is an important and unresolved question. Incorporating socioeconomic data as a cost We limited our review to costs that influence the spatial location of conservation areas within the systematic conservation planning framework. That is, we did not include nonspatially explicit costs. We used Web of Knowledge® ( http://www.isiknowledge.com ) to search for studies that applied spatially explicit decision support tools in the marine environment, using the following key phrases: (marine and conservation) and (“decision support tool” or Marxan or C‐Plan or Zonation or Resnet). We excluded studies that did not use costs and studies that did not specify the costs used. We then followed references from relevant studies to find additional peer‐reviewed and grey literature sources. Four approaches of using socioeconomic data in marine systematic prioritization exercises emerged from the literature ( Figure 2 ): (1) uniform cost or area as a proxy for human use (Table S1), (2) fisheries as the cost (Table S2), (3) multiple socioeconomic costs (Table S3), and (4) measures of naturalness or ecological impact of human activities (Table S4). Costs in marine systematic planning focused principally on representing opportunity costs. 2 Summary of the number of studies encountered in this review that use costs to influence site selection in the peer‐reviewed and grey literature. Uniform cost or area Prior to 2004, most planners used homogenous costs (e.g., area) ( Figure 3 ) ( Beck & Odaya 2001 ; Airamé 2003 ; Stewart 2003 ). The assumption (although not always stated) is that by minimizing the area, the impacts of conservation areas on people—opportunity costs—will be reduced ( Beck & Odaya 2001 ; Stewart 2003 ). Other reasons for using a uniform cost are that other aspects of the project were considering socioeconomics ( Airamé 2003 ) and that the planning exercise was intended for demonstration or academic purposes ( Cook & Auster 2005 ; Stewart 2007 ; Leathwick 2008 ). 3 Number of studies encountered in this review by type of socioeconomic cost used over time. The first study encountered that uses costs in systematic marine conservation planning was published in 1999. The number of studies has increased over time, and fisheries as a cost accounted for the largest increase. *2009 data are current up to August 24. Using a uniform cost is the simplest—albeit crude and perhaps misleading (see Figure 1 )—approach for giving consideration to socioeconomic factors ( Polasky 2005 ; Carwardine 2008b ). We speculate that the reason uniform costs dominated until recently ( Figure 3 ) is that (1) it is easy to generate while data for other costs can be difficult to obtain or unavailable, (2) the influence of using other costs was not recognized until recently due to research advances, and (3) planners were interested primarily in ecological aspects of conservation planning. In reality, conservation costs are spatially variable, making uniform costs a poor surrogate ( Naidoo 2006 ) ( Figure 1 ). For example, the cost of enforcing an MPA close to a port will be less than an equivalent‐sized reserve farther away (i.e., fuel, wages). Uniform costs result in a greater impact on marine activities than using more specific cost measures ( Stewart & Possingham 2005 ; Klein 2008a ). Therefore, if good quality information exists on the distribution of activities, it is preferable to include this information explicitly. Fisheries Fishing is the most prevalent human activity in the marine environment ( Pauly 2002 ). Most systematic conservation planning projects that used a spatially explicit socioeconomic cost used fisheries as the opportunity cost ( Figure 2 ). The catch or effort per unit area for a single fishery and multiple fisheries typically define the cost ( Richardson 2006 ; Lombard 2007 ; Klein 2008a, 2008b ). These metrics reflect the relative value of areas to the fisheries. Some studies combine catches with their monetary value (e.g. , ex‐vessel prices) ( Wood & Dragicevic 2007 ; Klein 2009 ). In some cases, the data indicate that fishing does not occur in some places. To ensure that these places are not “free” to conserve, the cost can be defined as a function of fishing costs and area ( Stewart & Possingham 2005 ; Game 2008 ). Although most studies focused on commercial fisheries, there are a few examples that considered other fishing sectors (e.g., recreational). Klein (2008a) developed a method for combining recreational and commercial fisheries relative to their value. Another study included artisanal fisheries, whereby the density of fishing boats per unit area was the proxy for small boat fishing pressure ( Sala 2002 ). Small‐scale or subsistence fisheries, especially in developing countries, are difficult to consider as they are rarely mapped ( Ban 2009a ). The advantage of using fisheries as the cost is that it minimizes the opportunity cost to a prevalent user group ( Klein 2008b ). When a single fishery would be affected by the conservation areas, that fishery can be used as the cost. When several fisheries would be affected, their uses can be combined to minimize the overall cost to all fisheries. The disadvantage is that combining multiple fisheries into one cost will not equitably impact each fishery ( Klein 2009 ). A similar effect occurs when combining multiple socioeconomic costs (see below). Multiple socioeconomic costs There are often other stakeholders (besides fishers) impacted by MPAs. For example, conservation actions might result in shipping lanes being re‐routed, mines being closed, and restrictions being placed on recreational uses. The challenge is to minimize the cost to such users appropriately. A logical yet challenging planning objective would be to minimize impacts to multiple users. There are few documented examples where different kinds of costs are combined ( Leathwick 2008 ; Green 2009 ). This may be because combining disparate socioeconomic data are often measured in different units (e.g. , number of oil tankers, fishing intensity, and aquaculture production) and/or can involve difficult decisions about weighting different costs. In the Great Barrier Reef (Australia) rezoning process, several different socioeconomic data are discussed but details of the approach are not well documented ( Lewis 2003 ; Fernandes 2005 ). In theory, the advantage of combining multiple socioeconomic data into one cost layer is that the overall impact on marine users can be minimized. In practice, combining disparate socioeconomic costs into one layer is challenging and a framework for overcoming this challenge does not exist. Even when such data can be combined, the overall impact is minimized but may not equitably impact user groups ( Klein 2009 ). This is the inherent challenge of having to combine multiple costs into one cost layer to be minimized. Naturalness or ecological impact of human activities Another “cost” used in marine planning is the ecological impact of human activities ( Banks 2005 ; Tallis 2008 ; Ban 2009b ). These studies use data that depict impacts of human activities on marine ecosystems, which is sometimes mapped using socioeconomic data; i.e., data sets depicting human uses are weighted based on their impact, then added to create an impact map ( Halpern 2008 ; Tallis 2008 ). The aim of this approach is to prioritize places for conservation that are impacted minimally by humans. All else equal, less impacted areas are prioritized over heavily impacted places. The disadvantage of this approach is that it does not minimize costs to specific user groups, and hence, is unlikely to result in MPAs that are cost‐effective to implement and manage. Human impact as a cost does not conform with the categories outlined by Naidoo . (2006) . Whether the use of this cost is appropriate depends on the specific objectives of the conservation plan. Other methods of combining socioeconomic data Different methods can be used to combine a variety of costs. Scoring systems, which assign points based on different parameters, have been criticized because combining scores is arbitrary and often not mathematically correct ( Possingham 2002 ; Mace et al. 2007). If based on expert input, the scoring approach can be less arbitrary and transparent if the process is properly documented (Mace et al. 2007). Multicriteria decision making (also referred to as multicriteria evaluation, multicriteria decision analysis, and multiple criterion synchronization) can be used to combine costs ( Moffett & Sarkar 2006 ; Sarkar 2006 ). Multicriteria decision making is usually applied to evaluate alternative scenarios of reserve networks based social, economic, political, and/or biological factors ( Moffett 2005 ). Software programs (e.g., MultCSync) can assists with this process ( Moffett 2005 ). Based on chosen criteria, the suitability of planning units as conservation areas or alternative network options are ranked ( Moffett & Sarkar 2006 ). Yet multicriteria decision making has rarely been used in the marine environment ( Brown 2001 ; Villa 2002 ; Wood & Dragicevic 2007 ). While the multicriteria decision‐making method has been recommended for use in decision support tools ( Moffett & Sarkar 2006 ), no examples of its use to combine costs were found. For a terrestrial analysis of different ways of integrating opportunity costs using multicriteria analysis, see Cameron (2008) . Combining costs in marine conservation planning Combining multiple costs into one cost is the primary challenge of incorporating socioeconomic data into conservation planning decision support tools. Any metric of costs can be used in decision support tools, as long as it addresses the objective of the conservation problem. The crucial step in identifying a sensible cost is to carefully formulate the conservation problem. The cost should then be chosen to directly relate to the objective. Combining many costs is feasible when each cost is measured in the same unit (e.g., dollars), but this is rarely the case in the marine environment (e.g. , CPUE and number of ships). Combining costs with different units can produce data that are not mathematically sensible, resulting in priority areas that do not achieve the stated objectives. Regardless of method used to create a single socioeconomic cost, it may not minimize impacts equitably to all users. This is a limiting factor of the mathematical formulation of the reserve selection problem as currently implemented in decision support tools. Marxan with Zones ( Watts 2009 ) a new version of the Marxan software, will improve our ability to accommodate multiple socioeconomic considerations in some conservation assessments. The addition of user‐defined zones and ability to specify costs and targets for each zone adds flexibility to balance biodiversity and socioeconomic objectives ( Watts 2009 ). An alternative is to use each socioeconomic data set as a cost (e.g. , each human use) in a separate scenario of the analysis ( Cameron 2008 ). For example, Klein (2008a) developed three scenarios, each with a different cost: (1) recreational fishers; (2) commercial fishers; and (3) recreational and commercial fishers combined. Using the Klein (2008a) results, we demonstrate three ways of using such an analysis to inform the design of protected areas: (1) The results of each scenario can be evaluated to determine their costs to each stakeholder group, clearly identifying trade‐offs ( Figure 4A ). (2) Statistical analyses (e.g., cluster analyses) can be used to assess the similarity of the results of scenarios (see Airame 2005 ), outlining similar solutions within and between scenarios which can identify the range of geospatial alternatives ( Figure 4B ). (3) A map showing the difference in the selection frequency between two scenarios is useful in comparing results between scenarios (see Carwardine 2008b ) ( Figure 4C ). 4 Examples of analyses and comparisons of conservation options for scenarios with different costs. (A) Percentage of recreational, commercial, and combined fishing effort displaced by marine protected areas identified in the best solutions generated by three different scenarios. Scenarios 1, 2, and 3 aimed to minimize the cost to the recreational, commercial, and combined recreational/commercial fishing industries, respectively. (B) Ten solutions from each of the recreational, commercial, and recreational/commercial fishing scenarios were sorted using a cluster analysis and grouped by similarity. R = recreational scenario; C = commercial scenario; RC = recreational and commercial combined scenario. (C) Difference in the spatial distribution of selection frequencies between protected‐area design scenarios 1 and 2. An advantage of this approach is that it would provide each user group with a scenario that directly relates to their interests, and allows researchers to assess the differences between human use sectors under the same conservation constraints. A user‐conflict analysis can then be carried out for the management objectives used in the scenarios. The disadvantage is that it would not create any potential reserve solutions that consider all costs. However, the results of combining multiple costs be used to test the sensitivity of priority areas to other cost scenarios. It will never be possible to include all socioeconomic considerations in a conservation planning tool. Decision support tools should only be used to support the design of MPAs given a clearly defined objective. Some socioeconomic considerations may be best considered post hoc or using other approaches, such as through planning tables involving stakeholder consensus building. Other observations and considerations Several themes are conspicuously absent from the systematic marine conservation planning literature. First, we encountered no examples of temporally and spatially dynamic socioeconomic data being used in systematic marine conservation planning. Second, we encountered no projections of spatially explicit future costs. Incorporating future uses would require the prediction of the spatial distribution (expansion or contraction) of relevant activities in the future ( Pressey 2007 ), so that siting algorithms can adequately represent user groups now and into the future when prioritizing areas for conservation. In this regard, marine planning lags behind its terrestrial counterpart, where future threat scenarios have been incorporated ( Sarkar 2006 ), although this is starting to be addresses, e.g., through the integration of ecosystem models and site selection algorithms (Christensen et al. , in press). Third, explicit integration of implementation opportunity is rare (but see Green 2009 ). For example, some communities may be amenable to having a protected area in their waters, others may not (see Knight 2006 for a terrestrial example). Finally, few examples exist of explicit trade‐off analyses—e.g. , trade‐offs between user groups, and between users and ecological objectives. While our review focused on incorporating socioeconomic data as a cost to be minimized within marine conservation planning tools, these data can be used in two other ways. First, if the inclusion of a socioeconomic feature is desired in an MPA (i.e., an area of cultural importance, such as a ship wreck), then it can be targeted in the same way as biodiversity features ( Klein 2008a ). Also, priority areas for fishing can be identified (rather than MPAs) and the fishing distributions can be targeted for inclusion ( Ban & Vincent 2009 ; Klein 2009 ). Second, planning units can be excluded or always included in the result. Some activities may have permanent infrastructure that is not feasible to include in protected areas, e.g. , oil rigs or areas where it is known that a community is not interested in engaging in conservation ( Green 2009 ). Similarly, planning units can always be included as conservation areas in the result, e.g., when stakeholders agree on a place to be protected. While not the focus of our review, socioeconomic data are also used in post‐hoc analyses of the anticipated effect of marine reserves. In particular, spatially explicit bioeconomic modeling of fisheries can assess the potential cost or benefit of protected areas to one or more fisheries ( Holland 2000 ; Smith & Wilen 2003 ). There is much scope for developing methods for including bioeconomic models in systematic conservation planning. Conclusion The framework developed for terrestrial systems can generally be applied to marine systems, but opportunity costs are relatively more important in marine assessments due to property rights and tenure. The most prominent costs to marine conservation areas are the opportunity costs to fishers. Recently, planners have shifted from using area as a surrogate for opportunity costs to using fishing data. Although this shift is a substantial improvement upon the way we assess conservation areas, developments still must be made to represent the true opportunity costs. To date, the cost used in marine conservation planning has been static, despite the dynamic nature of fishing. Static fishing costs do not represent the true cost to fishers and could be improved by considering the temporal variation of fishing costs, benefits of spillover, and redistribution of effort after reservation. This is an important area of further research, albeit one that requires substantial amounts of information on fleet behavior, fish populations, and other dynamic parameters ( Pelletier & Mahevas 2005 , Branch 2006 ). Several recommendations emerge that would present a much‐needed shift in the way costs are represented in conservation planning exercises. A clear definition of the objectives, including the socioeconomic objectives and metrics, is essential for meaningful systematic conservation planning. Arbitrary combinations of costs should be avoided if they cannot be adequately defended. The rationale and methods of combining costs need to be clearly documented. When multiple socioeconomic data sets are important as costs, analyses that consider each cost separately may be more appropriate than an arbitrary combination of such costs. Different ways of combining multiple costs can be used to test the sensitivity of the decision support tool to the cost. Yet it will never be possible to incorporate everything into a marine plan or a tool. There is much scope for improved use of socioeconomic costs in systematic marine conservation planning. Research into options for combining disparate costs is needed. Integrating temporal and spatial dynamics of costs, future cost projections, and implementation opportunity lie at the heart of the challenge to forge realistic marine conservation plans. Editor : Prof. Stephen Polasky Acknowledgments This review was instigated as part of the British Columbia Marine Conservation Analysis (BCMCA) project, funded by the Gordon and Betty Moore Foundation, David Suzuki Foundation, the Province of British Columbia (ILMB), David and Lucile Packard Foundation, the Pacific Marine Analysis and Research Association, Living Ocean Society, and in‐kind staff and resource contributions of participating organizations. We thank V. Adams, K. Bodtker, T. Bryan, N. Davis, A. Day, E. Game, K. Royle, D. Segan, and C. Short for insightful comments on previous versions of the manuscript. We also thank the editors of Conservation Letters and two anonymous reviewers for greatly improving the manuscript. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Conservation Letters Wiley

Spatial socioeconomic data as a cost in systematic marine conservation planning

Loading next page...
 
/lp/wiley/spatial-socioeconomic-data-as-a-cost-in-systematic-marine-conservation-qp7grt7kfW

References (64)

Publisher
Wiley
Copyright
"Copyright © 2009 Wiley Subscription Services, Inc., A Wiley Company"
eISSN
1755-263X
DOI
10.1111/j.1755-263X.2009.00071.x
Publisher site
See Article on Publisher Site

Abstract

Introduction In planning for conservation, planners and scientists have focused on the biological benefits of conservation plans (i.e., Naidoo 2006 ). There is increasing recognition that conservation objectives must be achieved efficiently because of limited conservation resources. Hence, socioeconomic costs must be integrated into conservation planning ( Naidoo 2006 ; Carwardine 2008b ), where “cost” is intended to reflect the socioeconomic impacts of conservation areas. There has been an influx of initiatives to promote the design and implementation of marine protected areas (MPAs) around the globe ( Spalding 2008 ) and many marine planners are seeking guidance on how to use socioeconomic data. By explicitly incorporating socioeconomic costs into systematic marine conservation planning, we can avoid costly conservation mistakes. Systematic conservation planning is an approach that guides the location and design of conservation areas that achieve explicit biodiversity objectives ( Margules & Pressey 2000 ). Application of this approach to real‐world conservation assessments is supported by decision support tools, e.g. , Marxan ( Ball & Possingham 2000 ; Possingham 2000 ), C‐Plan ( Pressey 2009 ), Zonation ( Moilanen 2005 ), and ResNet ( Kelley 2002 ). These tools were designed to support, not make, decisions on the location of conservation areas; they provide the basis for discussions but do not provide an answer that is to be unequivocally accepted ( Possingham 2006 ). A common problem planners aim to solve is to minimize the cost of conservation areas while achieving quantitative representation targets ( Airamé 2003 ; Stewart & Possingham 2005 ; Naidoo 2006 ; Possingham 2006 ; Klein 2008b ). The Marxan software was designed to solve this problem. Recent literature highlights the importance of incorporating socioeconomic costs into conservation planning for two main reasons. First, including socioeconomic costs minimizes the impacts on resource users ( Klein 2008a, 2008b ) and thus reduces conflicts between resource users and conservationists ( Scholz 2004 ; Crawford 2006 ; Carwardine 2008b ; Klein 2008a ). Second, including such costs produces plans that are cost‐effective to implement and manage ( Naidoo 2006 ; Carwardine 2008b ). We summarize the costs of marine conservation, review how these costs have been used in conservation planning, and discuss future research priorities to address shortcomings of current approaches. Reviews on incorporating costs into conservation planning have been conducted for terrestrial planning ( Hughey 2003 ; O’Connor 2003 ; Polasky 2005 ; Naidoo 2006 ). A review highlighting the unique aspects of socioeconomic costs in marine planning is necessary because: (1) aspects of marine systems differ from their terrestrial counterpart (see below); (2) the field has progressed quickly over a short time; and (3) many practical applications of these concepts are difficult to find in the literature yet useful and sought after by marine planners. Marine conservation costs Many costs are associated with conservation planning, including acquisition costs, management costs, transaction costs, and opportunity costs (see Naidoo 2006 ). Acquisition costs are atypical in the marine environment because waters are not usually privately owned. Management costs are associated with enforcing and maintaining MPAs ( Naidoo 2006 ). Transaction costs are associated with negotiating protection, such as the time and staff involved in stakeholder negotiations. Opportunity costs are foregone revenues, e.g., the value to fisheries and other marine uses ( Naidoo 2006 ), and are commonly used to site MPAs. Damage costs are associated with loss to economic activities due to conservation activities; an example from land is elephants or monkeys from reserves raiding agricultural fields ( Naidoo 2006 ). The relative importance of these types of costs differs in terrestrial and marine systems. For example, the location of MPAs is predominantly influenced by the opportunity cost to industries (e.g. , fishing, mining). Although opportunity costs are also important in terrestrial planning, an additional cost is associated with the land tenure (e.g. , acquisition or stewardship costs). Land tenure or property rights systems influence the types of costs that predominate. Marine systems are typically common‐pool resources ( Ostrom 1999 ), whereas terrestrial systems have a much higher proportion of private ownership, although exceptions exist (e.g. , Beck 2004 ). Therefore, acquisition costs equivalent to the purchase of land are rare in the marine environment. Management and transaction costs are likely to be of similar importance, but little information exists on such costs ( Balmford 2004 ). We have not encountered any mention of damage costs due to marine conservation. Critical to identifying costs for MPAs is clearly defining the objective and specifying feasible conservation actions (e.g., no‐take marine reserve, multi‐purpose conservation area) ( Carwardine 2008a ). For example, if the objective is to select MPAs while minimizing the effect of such closures on commercial fishers, then the most suitable opportunity cost would be a measure of their impact on these fisheries. The cost impacts the spatial location of MPAs ( Figure 1 ). Using inappropriate cost measures can lead to costly conservation mistakes ( Carwardine 2008b ). For example, Carwardine (2008b) found that biodiversity targets can be up to twice as expensive to achieve if incorrect cost data are used. The costs identified vis‐à‐vis the objective have to be summarized into a single cost unit in one layer because the current formulation used in decision‐support tools (e.g., Marxan, C‐Plan, Resnet) allows for only one cost layer. 1 Results of scenarios using different costs. The left column shows the importance to conservation of areas under a variety of cost scenarios, and the right column depicts the corresponding costs: (a) area as a cost, (b) human impacts as a cost, and (c) commercial fisheries as a cost. The maps depict scenarios that represent 10% of every feature. Reproduced from Ban (2009b) . For use in planning, cost data have to be spatially explicit at a scale fine enough to differentiate areas. When spatially explicit cost data are not available, surrogates can be used (e.g., global fishing effort, or coastal population density as a proxy of fishing effort, Ban 2009a ). Even when spatially explicit costs are available, these often represent surrogates for the real costs of conservation planning. For example, a marine plan might want to minimize the monetary cost of establishing MPAs, but uses areas of relative importance to commercial fisheries as the surrogate. How well such surrogates represent actual cost efficiencies is an important and unresolved question. Incorporating socioeconomic data as a cost We limited our review to costs that influence the spatial location of conservation areas within the systematic conservation planning framework. That is, we did not include nonspatially explicit costs. We used Web of Knowledge® ( http://www.isiknowledge.com ) to search for studies that applied spatially explicit decision support tools in the marine environment, using the following key phrases: (marine and conservation) and (“decision support tool” or Marxan or C‐Plan or Zonation or Resnet). We excluded studies that did not use costs and studies that did not specify the costs used. We then followed references from relevant studies to find additional peer‐reviewed and grey literature sources. Four approaches of using socioeconomic data in marine systematic prioritization exercises emerged from the literature ( Figure 2 ): (1) uniform cost or area as a proxy for human use (Table S1), (2) fisheries as the cost (Table S2), (3) multiple socioeconomic costs (Table S3), and (4) measures of naturalness or ecological impact of human activities (Table S4). Costs in marine systematic planning focused principally on representing opportunity costs. 2 Summary of the number of studies encountered in this review that use costs to influence site selection in the peer‐reviewed and grey literature. Uniform cost or area Prior to 2004, most planners used homogenous costs (e.g., area) ( Figure 3 ) ( Beck & Odaya 2001 ; Airamé 2003 ; Stewart 2003 ). The assumption (although not always stated) is that by minimizing the area, the impacts of conservation areas on people—opportunity costs—will be reduced ( Beck & Odaya 2001 ; Stewart 2003 ). Other reasons for using a uniform cost are that other aspects of the project were considering socioeconomics ( Airamé 2003 ) and that the planning exercise was intended for demonstration or academic purposes ( Cook & Auster 2005 ; Stewart 2007 ; Leathwick 2008 ). 3 Number of studies encountered in this review by type of socioeconomic cost used over time. The first study encountered that uses costs in systematic marine conservation planning was published in 1999. The number of studies has increased over time, and fisheries as a cost accounted for the largest increase. *2009 data are current up to August 24. Using a uniform cost is the simplest—albeit crude and perhaps misleading (see Figure 1 )—approach for giving consideration to socioeconomic factors ( Polasky 2005 ; Carwardine 2008b ). We speculate that the reason uniform costs dominated until recently ( Figure 3 ) is that (1) it is easy to generate while data for other costs can be difficult to obtain or unavailable, (2) the influence of using other costs was not recognized until recently due to research advances, and (3) planners were interested primarily in ecological aspects of conservation planning. In reality, conservation costs are spatially variable, making uniform costs a poor surrogate ( Naidoo 2006 ) ( Figure 1 ). For example, the cost of enforcing an MPA close to a port will be less than an equivalent‐sized reserve farther away (i.e., fuel, wages). Uniform costs result in a greater impact on marine activities than using more specific cost measures ( Stewart & Possingham 2005 ; Klein 2008a ). Therefore, if good quality information exists on the distribution of activities, it is preferable to include this information explicitly. Fisheries Fishing is the most prevalent human activity in the marine environment ( Pauly 2002 ). Most systematic conservation planning projects that used a spatially explicit socioeconomic cost used fisheries as the opportunity cost ( Figure 2 ). The catch or effort per unit area for a single fishery and multiple fisheries typically define the cost ( Richardson 2006 ; Lombard 2007 ; Klein 2008a, 2008b ). These metrics reflect the relative value of areas to the fisheries. Some studies combine catches with their monetary value (e.g. , ex‐vessel prices) ( Wood & Dragicevic 2007 ; Klein 2009 ). In some cases, the data indicate that fishing does not occur in some places. To ensure that these places are not “free” to conserve, the cost can be defined as a function of fishing costs and area ( Stewart & Possingham 2005 ; Game 2008 ). Although most studies focused on commercial fisheries, there are a few examples that considered other fishing sectors (e.g., recreational). Klein (2008a) developed a method for combining recreational and commercial fisheries relative to their value. Another study included artisanal fisheries, whereby the density of fishing boats per unit area was the proxy for small boat fishing pressure ( Sala 2002 ). Small‐scale or subsistence fisheries, especially in developing countries, are difficult to consider as they are rarely mapped ( Ban 2009a ). The advantage of using fisheries as the cost is that it minimizes the opportunity cost to a prevalent user group ( Klein 2008b ). When a single fishery would be affected by the conservation areas, that fishery can be used as the cost. When several fisheries would be affected, their uses can be combined to minimize the overall cost to all fisheries. The disadvantage is that combining multiple fisheries into one cost will not equitably impact each fishery ( Klein 2009 ). A similar effect occurs when combining multiple socioeconomic costs (see below). Multiple socioeconomic costs There are often other stakeholders (besides fishers) impacted by MPAs. For example, conservation actions might result in shipping lanes being re‐routed, mines being closed, and restrictions being placed on recreational uses. The challenge is to minimize the cost to such users appropriately. A logical yet challenging planning objective would be to minimize impacts to multiple users. There are few documented examples where different kinds of costs are combined ( Leathwick 2008 ; Green 2009 ). This may be because combining disparate socioeconomic data are often measured in different units (e.g. , number of oil tankers, fishing intensity, and aquaculture production) and/or can involve difficult decisions about weighting different costs. In the Great Barrier Reef (Australia) rezoning process, several different socioeconomic data are discussed but details of the approach are not well documented ( Lewis 2003 ; Fernandes 2005 ). In theory, the advantage of combining multiple socioeconomic data into one cost layer is that the overall impact on marine users can be minimized. In practice, combining disparate socioeconomic costs into one layer is challenging and a framework for overcoming this challenge does not exist. Even when such data can be combined, the overall impact is minimized but may not equitably impact user groups ( Klein 2009 ). This is the inherent challenge of having to combine multiple costs into one cost layer to be minimized. Naturalness or ecological impact of human activities Another “cost” used in marine planning is the ecological impact of human activities ( Banks 2005 ; Tallis 2008 ; Ban 2009b ). These studies use data that depict impacts of human activities on marine ecosystems, which is sometimes mapped using socioeconomic data; i.e., data sets depicting human uses are weighted based on their impact, then added to create an impact map ( Halpern 2008 ; Tallis 2008 ). The aim of this approach is to prioritize places for conservation that are impacted minimally by humans. All else equal, less impacted areas are prioritized over heavily impacted places. The disadvantage of this approach is that it does not minimize costs to specific user groups, and hence, is unlikely to result in MPAs that are cost‐effective to implement and manage. Human impact as a cost does not conform with the categories outlined by Naidoo . (2006) . Whether the use of this cost is appropriate depends on the specific objectives of the conservation plan. Other methods of combining socioeconomic data Different methods can be used to combine a variety of costs. Scoring systems, which assign points based on different parameters, have been criticized because combining scores is arbitrary and often not mathematically correct ( Possingham 2002 ; Mace et al. 2007). If based on expert input, the scoring approach can be less arbitrary and transparent if the process is properly documented (Mace et al. 2007). Multicriteria decision making (also referred to as multicriteria evaluation, multicriteria decision analysis, and multiple criterion synchronization) can be used to combine costs ( Moffett & Sarkar 2006 ; Sarkar 2006 ). Multicriteria decision making is usually applied to evaluate alternative scenarios of reserve networks based social, economic, political, and/or biological factors ( Moffett 2005 ). Software programs (e.g., MultCSync) can assists with this process ( Moffett 2005 ). Based on chosen criteria, the suitability of planning units as conservation areas or alternative network options are ranked ( Moffett & Sarkar 2006 ). Yet multicriteria decision making has rarely been used in the marine environment ( Brown 2001 ; Villa 2002 ; Wood & Dragicevic 2007 ). While the multicriteria decision‐making method has been recommended for use in decision support tools ( Moffett & Sarkar 2006 ), no examples of its use to combine costs were found. For a terrestrial analysis of different ways of integrating opportunity costs using multicriteria analysis, see Cameron (2008) . Combining costs in marine conservation planning Combining multiple costs into one cost is the primary challenge of incorporating socioeconomic data into conservation planning decision support tools. Any metric of costs can be used in decision support tools, as long as it addresses the objective of the conservation problem. The crucial step in identifying a sensible cost is to carefully formulate the conservation problem. The cost should then be chosen to directly relate to the objective. Combining many costs is feasible when each cost is measured in the same unit (e.g., dollars), but this is rarely the case in the marine environment (e.g. , CPUE and number of ships). Combining costs with different units can produce data that are not mathematically sensible, resulting in priority areas that do not achieve the stated objectives. Regardless of method used to create a single socioeconomic cost, it may not minimize impacts equitably to all users. This is a limiting factor of the mathematical formulation of the reserve selection problem as currently implemented in decision support tools. Marxan with Zones ( Watts 2009 ) a new version of the Marxan software, will improve our ability to accommodate multiple socioeconomic considerations in some conservation assessments. The addition of user‐defined zones and ability to specify costs and targets for each zone adds flexibility to balance biodiversity and socioeconomic objectives ( Watts 2009 ). An alternative is to use each socioeconomic data set as a cost (e.g. , each human use) in a separate scenario of the analysis ( Cameron 2008 ). For example, Klein (2008a) developed three scenarios, each with a different cost: (1) recreational fishers; (2) commercial fishers; and (3) recreational and commercial fishers combined. Using the Klein (2008a) results, we demonstrate three ways of using such an analysis to inform the design of protected areas: (1) The results of each scenario can be evaluated to determine their costs to each stakeholder group, clearly identifying trade‐offs ( Figure 4A ). (2) Statistical analyses (e.g., cluster analyses) can be used to assess the similarity of the results of scenarios (see Airame 2005 ), outlining similar solutions within and between scenarios which can identify the range of geospatial alternatives ( Figure 4B ). (3) A map showing the difference in the selection frequency between two scenarios is useful in comparing results between scenarios (see Carwardine 2008b ) ( Figure 4C ). 4 Examples of analyses and comparisons of conservation options for scenarios with different costs. (A) Percentage of recreational, commercial, and combined fishing effort displaced by marine protected areas identified in the best solutions generated by three different scenarios. Scenarios 1, 2, and 3 aimed to minimize the cost to the recreational, commercial, and combined recreational/commercial fishing industries, respectively. (B) Ten solutions from each of the recreational, commercial, and recreational/commercial fishing scenarios were sorted using a cluster analysis and grouped by similarity. R = recreational scenario; C = commercial scenario; RC = recreational and commercial combined scenario. (C) Difference in the spatial distribution of selection frequencies between protected‐area design scenarios 1 and 2. An advantage of this approach is that it would provide each user group with a scenario that directly relates to their interests, and allows researchers to assess the differences between human use sectors under the same conservation constraints. A user‐conflict analysis can then be carried out for the management objectives used in the scenarios. The disadvantage is that it would not create any potential reserve solutions that consider all costs. However, the results of combining multiple costs be used to test the sensitivity of priority areas to other cost scenarios. It will never be possible to include all socioeconomic considerations in a conservation planning tool. Decision support tools should only be used to support the design of MPAs given a clearly defined objective. Some socioeconomic considerations may be best considered post hoc or using other approaches, such as through planning tables involving stakeholder consensus building. Other observations and considerations Several themes are conspicuously absent from the systematic marine conservation planning literature. First, we encountered no examples of temporally and spatially dynamic socioeconomic data being used in systematic marine conservation planning. Second, we encountered no projections of spatially explicit future costs. Incorporating future uses would require the prediction of the spatial distribution (expansion or contraction) of relevant activities in the future ( Pressey 2007 ), so that siting algorithms can adequately represent user groups now and into the future when prioritizing areas for conservation. In this regard, marine planning lags behind its terrestrial counterpart, where future threat scenarios have been incorporated ( Sarkar 2006 ), although this is starting to be addresses, e.g., through the integration of ecosystem models and site selection algorithms (Christensen et al. , in press). Third, explicit integration of implementation opportunity is rare (but see Green 2009 ). For example, some communities may be amenable to having a protected area in their waters, others may not (see Knight 2006 for a terrestrial example). Finally, few examples exist of explicit trade‐off analyses—e.g. , trade‐offs between user groups, and between users and ecological objectives. While our review focused on incorporating socioeconomic data as a cost to be minimized within marine conservation planning tools, these data can be used in two other ways. First, if the inclusion of a socioeconomic feature is desired in an MPA (i.e., an area of cultural importance, such as a ship wreck), then it can be targeted in the same way as biodiversity features ( Klein 2008a ). Also, priority areas for fishing can be identified (rather than MPAs) and the fishing distributions can be targeted for inclusion ( Ban & Vincent 2009 ; Klein 2009 ). Second, planning units can be excluded or always included in the result. Some activities may have permanent infrastructure that is not feasible to include in protected areas, e.g. , oil rigs or areas where it is known that a community is not interested in engaging in conservation ( Green 2009 ). Similarly, planning units can always be included as conservation areas in the result, e.g., when stakeholders agree on a place to be protected. While not the focus of our review, socioeconomic data are also used in post‐hoc analyses of the anticipated effect of marine reserves. In particular, spatially explicit bioeconomic modeling of fisheries can assess the potential cost or benefit of protected areas to one or more fisheries ( Holland 2000 ; Smith & Wilen 2003 ). There is much scope for developing methods for including bioeconomic models in systematic conservation planning. Conclusion The framework developed for terrestrial systems can generally be applied to marine systems, but opportunity costs are relatively more important in marine assessments due to property rights and tenure. The most prominent costs to marine conservation areas are the opportunity costs to fishers. Recently, planners have shifted from using area as a surrogate for opportunity costs to using fishing data. Although this shift is a substantial improvement upon the way we assess conservation areas, developments still must be made to represent the true opportunity costs. To date, the cost used in marine conservation planning has been static, despite the dynamic nature of fishing. Static fishing costs do not represent the true cost to fishers and could be improved by considering the temporal variation of fishing costs, benefits of spillover, and redistribution of effort after reservation. This is an important area of further research, albeit one that requires substantial amounts of information on fleet behavior, fish populations, and other dynamic parameters ( Pelletier & Mahevas 2005 , Branch 2006 ). Several recommendations emerge that would present a much‐needed shift in the way costs are represented in conservation planning exercises. A clear definition of the objectives, including the socioeconomic objectives and metrics, is essential for meaningful systematic conservation planning. Arbitrary combinations of costs should be avoided if they cannot be adequately defended. The rationale and methods of combining costs need to be clearly documented. When multiple socioeconomic data sets are important as costs, analyses that consider each cost separately may be more appropriate than an arbitrary combination of such costs. Different ways of combining multiple costs can be used to test the sensitivity of the decision support tool to the cost. Yet it will never be possible to incorporate everything into a marine plan or a tool. There is much scope for improved use of socioeconomic costs in systematic marine conservation planning. Research into options for combining disparate costs is needed. Integrating temporal and spatial dynamics of costs, future cost projections, and implementation opportunity lie at the heart of the challenge to forge realistic marine conservation plans. Editor : Prof. Stephen Polasky Acknowledgments This review was instigated as part of the British Columbia Marine Conservation Analysis (BCMCA) project, funded by the Gordon and Betty Moore Foundation, David Suzuki Foundation, the Province of British Columbia (ILMB), David and Lucile Packard Foundation, the Pacific Marine Analysis and Research Association, Living Ocean Society, and in‐kind staff and resource contributions of participating organizations. We thank V. Adams, K. Bodtker, T. Bryan, N. Davis, A. Day, E. Game, K. Royle, D. Segan, and C. Short for insightful comments on previous versions of the manuscript. We also thank the editors of Conservation Letters and two anonymous reviewers for greatly improving the manuscript.

Journal

Conservation LettersWiley

Published: Oct 1, 2009

Keywords: ; ; ; ; ; ;

There are no references for this article.