Correspondence Edward T. Game, The Nature Conservancy, South Brisbane, Queensland 4101, Australia. Tel: +61-7-3214-6921; fax: +61-7-3214-6999. E-mail: firstname.lastname@example.org
Human factors more than ecology dictate conservation opportunity and the subsequent success of implementation. This is particularly true in places such as the Solomon Islands where most terrestrial and coastal marine areas remain in community ownership. However, factors such as community support are not reliably predictable, nor easy to map, and therefore challenging to incorporate into systematic conservation plans. Here, we describe how the Lauru Land Conference of Tribal Communities and The Nature Conservancy have worked with the communities of Choiseul Province, Solomon Islands, to develop a conservation planning process that reconciles community-driven conservation opportunities, with a systematic and representation-based approach to prioritization. We suggest how sophisticated prioritization software can be used collaboratively in a community setting, to dynamically assess and guide conservation opportunities as they arise; a process of informed opportunism.
One response to this problem is for conservation assessments to map and incorporate a range of social, economic, and political factors that are likely to define conservation cost and/or opportunity in a region (Knight & Cowling 2007). While there is little doubt that accounting for human opportunities and constraints as part of the planning process will improve the feasibility for implementation of proposed conservation sites, it also introduces its own set of challenges. Factors, such as community support, are difficult to quantify and map (but see, Knight et al. 2010) and can fluctuate both with the profitability of different land- and sea-use activities (Wunder 2007) and as the composition of communities change. The necessary data can also be time consuming to collect. While the inclusion of social, economic, and political data in conservation assessments is likely to make conservation recommendations more aligned with opportunities, it does not make conservation truly responsive to opportunities (Meir et al. 2004; Pressey & Bottrill 2008).
An alternative to trying to predict the distribution of conservation opportunities is for systematic conservation planning to be used more dynamically to assess and guide conservation opportunities as they arise; a sort of informed opportunism (Noss et al. 2002). Unfortunately, there is a common view among many conservation practitioners that adherence to priorities identified through systematic conservation assessments can prevent conservation capitalizing on real opportunities for action. Systematic conservation planning arose not as the antithesis of opportunistic selection of protected areas, but to counter biased selection toward areas of limited commercial value (Pressey 1994). This important distinction is too often overlooked. Over the last decade, systematic conservation planning has increasingly promoted dynamism, iteration, and stakeholder consultation as key features (e.g., Costello & Polasky 2004; Knight et al. 2006; Pressey & Bottrill 2008), such that informed opportunism should really be considered standard practice.
In reality, however, systematic conservation assessments have rarely been as dynamic and iterative as intended. The effort invested to develop a single “solution” has often been so great that there is little appetite for ongoing refinement (Meir et al. 2004). This is commonly compounded by a lack of technical expertise within the implementing agency, to manipulate the data and software required for a systematic assessment. Because of these and other challenges, conservation plans can quickly become obsolete, less relevant to conservation decision making, and therefore there is even less impetus to update them as conditions change.
The issue of opportunism in conservation is particularly critical in the countries of Melanesia. Around 97% of the land in Papua New Guinea and the Solomon Islands is still under traditional ownership (Mullins & Flaherty 1995; Brown & Durst 2003), and conservation action is simply not possible without strong landowner support. Top down prioritizations have proven ineffectual at catalyzing conservation action in Melanesia, largely because the subsequent negotiations with landowners have often proven tricky, expensive, and ultimately have not been adhered to because of a lack of genuine support (Rose Singadan, Department of Environment and Conservation, Government of Papua New Guinea, pers. comm.). Perversely, the a priori identification of conservation priorities has meant that some communities who are willing and ready to engage in conservation actions, and whose lands and waters are suitable for conservation, have not been adequately supported (Richard Hamilton, The Nature Conservancy, pers. obs.).
Choiseul (Fig. 1) is the most biodiverse island in the Solomon Archipelago (Diamond & Mayr 1976; Morrison et al. 2007; Keppel et al. 2010), and as well as being part of the Coral Triangle, the global center of coral reef diversity (Veron et al. 2009), contains some of the largest remaining stands of lowland rainforest in the Pacific (McClatchey et al. 2005). The Nature Conservancy has, for 9 years, worked closely with the Lauru Land Conference of Tribal Communities to build bottom-up support for a network of locally implemented and managed protected areas in Choiseul Province, Solomon Islands. Through this partnership, nine community-based protected areas have already been established on Choiseul, and by 2008 there was a ground swell of interest and enthusiasm among the communities of Choiseul for establishing both marine and terrestrial protected areas. Community-based protected areas on Choiseul are established as strict no-take areas, with harvest prohibition managed entirely by the customary owners of the site.
Conservation capacity within the Lauru Land Conference of Tribal Communities remains small, and by necessity the network of protected areas must be developed sequentially. The challenge this presents is to reconcile bottom-up, community-driven conservation opportunities with The Nature Conservancy's systematic and representation-based approach to prioritization (Groves et al. 2002). In other words, how to ensure the whole is more than the sum of the parts.
Here, we describe how the Lauru Land Conference of Tribal Communities and The Nature Conservancy have worked with the communities of Choiseul to develop a conservation planning process that address this challenge. Importantly, we demonstrate how a sophisticated spatial prioritization tool can be used in a collaborative fashion to support a locally driven process of informed opportunism.
Community engagement and institutional relationships
In 2000, The Nature Conservancy formed a partnership with the Lauru Land Conference of Tribal Communities. Initially, this partnership involved staff from The Nature Conservancy attending the Lauru Land Conference of Tribal Communities’ annual meeting and giving awareness-raising environmental presentations. In 2005, this partnership resulted in the formation of an Environmental Committee of the Lauru Land Conference of Tribal Communities, employment of a full time Environmental Community Conservation Officer, and establishment of a permanent environmental office in Choiseul. The Environmental Community Conservation Officer is the primary point of contact for local communities interested in establishing protected areas and played a critical role introducing and promoting the concept of a network of conservation areas. For the purposes of establishing a protected area, each community is represented by a small group of leaders from that community, typically comprising of chiefs, elders, and primary landowners. Before these leaders approach the Lauru Land Conference of Tribal Communities about establishing a protected area, a full community meeting is held to ensure consensus is reached and there are no unresolved conflicts over landownership in the proposed area. One of the first tasks of the Environmental Community Conservation Officer is to validate this consensus. These same community leaders serve as the primary point of contact for subsequent engagement and negotiation regarding the protected area.
In 2008, the large number of protected area opportunities being presented to the Lauru Land Conference of Tribal Communities through the Environmental Community Conservation Officer, provided a catalyst for future protected areas to be established in a more strategic manner in order to protect the full spectrum of Choiseul's biodiversity and natural resources. To initiate this process, The Nature Conservancy, the Lauru Land Conference of Tribal Communities, and the Choiseul Provincial Government convened a conservation planning workshop in May 2009. This workshop brought together community members from each of Choiseul's 12 wards. The workshop had four main aims: (1) to present the concept of a representative protected area network and introduce the kinds of data that could support its establishment; (2) to help validate the existing base biodiversity data through local knowledge; (3) collect spatial data on additional conservation features important to communities, as well as the location of areas traditionally managed for resource protection; (4) to discuss a process for identifying and establishing future protected areas, and how spatial prioritization software could support this process. Details of the planning region and base data used in the workshop can be found in Appendix A.
Aims 2 and 3 were supported through a process of participatory mapping. Large format (1:70,000), color maps illustrating existing base data on terrestrial (vegetation) and marine (reef) biodiversity were provided (see Appendix A), and participants from each ward were asked to delineate and label areas or sites of important natural resources (Fig. 2). Aim 4 was approached by identifying the key decisions that needed to be made regarding where to site a new protected area, who makes these decisions, and what information would be needed to support the decisions.
Following the workshop in May 2009, the community-edited maps were returned to Brisbane so all community added features could be digitized and added to the database. At the annual meeting of the Lauru Land Conference of Tribal Communities in October 2009, The Nature Conservancy staff from Brisbane worked with the Environmental Community Conservation Officer to present the complete set of conservation features on Choiseul, identify important features and resources not represented within the current set of protected areas, and develop with the communities a potential process for using the data collected to inform the establishment of a locally driven but representative network of protected areas. In addition, a firm, clear, and community endorsed agreement was sought on overall conservation objectives and quantitative conservation targets for Choiseul.
As a result of the participatory mapping exercise, an additional 25 conservation features were identified and mapped by the communities of Choiseul. These features represented important biological and cultural resources that would benefit from protection but were not captured by the base data layers, such as turtle nesting beaches, fish spawning aggregations, megapode nesting areas, and other aquatic resources. The mapping also revealed a number of sites that are already traditionally managed for resource protection. The only changes to the base data layers were the identification of areas where logging had occurred but was either not captured or updated in the underlying vegetation and logging history data.
Objectives and targets
At the Lauru Land Conference of Tribal Communities’ annual meeting in October 2009, the communities of Choiseul unanimously agreed to support two recommendations put forward jointly by the Environmental Committee of Lauru Land Conference of Tribal Communities and The Nature Conservancy: (1) establish a community-based representative protected areas network for the province of Choiseul; (2) designate at least one marine and one terrestrial protected area in each of the island's 12 wards by the end of 2012.
It was agreed that the target representation of conservation features in the protected area network would be set at 10% of the total area of each feature, in line with the Solomon Islands’ commitment under the Convention of Biological Diversity (CBD 2006). For important food resources such as fish spawning aggregations, and for threatened species such as turtle nesting colonies, the representation target was raised to 50% of the total area of those features. This was a community chosen target, based on the belief that 10% protection was inadequate to conserve these culturally valuable, but also threatened, ecological assets.
Site selection approach
Community support for a representative network of protected areas depended on having a mechanism that could efficiently identify important and complementary sites, but yet allowed the ultimate selection and implementation of protected areas to remain a flexible and community-driven process. When establishing new protected area sites, two critical decisions were identified: First, which communities were priorities for conservation action? And second, what site within a communities’ area of tenure would be protected? To support these decisions, a process was developed based on the use of Zonae Cogito (Watts et al. 2009), a new interface for the spatial prioritization software Marxan (Ball et al. 2009).
First, when the Lauru Land Conference of Tribal Communities and The Nature Conservancy receive multiple expressions of interest from Choiseul communities keen to establish a protected area in their territory, the selection frequency output, a measure of how many times each planning unit is selected across multiple runs, can be used to assess which communities’ territorial lands and seas contain the most “irreplaceable” sites based on representing terrestrial and marine habitats, as well as the locally identified conservation features (see, Carwardine et al. 2007; Ball et al. 2009). Engagement with communities can be prioritized on this basis, with decisions being made by the Environmental Committee of Lauru Land Conference of Tribal Communities and The Nature Conservancy.
Communities felt strongly that existing interest in conservation should be recognized in this initial prioritization. To accommodate this, the cost layer used in Marxan was the area of each planning unit, minus the proportion of the planning unit with conditions highly conducive to rapid inclusion in the protected area network. These included sites that are proposed but not yet gazetted protected areas, sites already managed by communities for natural resources, and sites where communities have previously indicated their support for the establishment of a protected area.
Second, once a decision is made to engage a community, a subsequent decision needs to be made, about which particular piece of land or sea will be set aside as a protected area within the community's area of tenure. This decision ultimately resides with the community. To ensure that opportunities for genuine community support are capitalized on, and yet protected area decisions still work systematically and efficiently toward a representative protected area network, a dynamic, interactive, and collaborative procedure was devised, making use of new capabilities in the Zonae Cogito software. The steps in this procedure are outlined in Table 1.
Table 1. The Zonae Cogito supported process of informed opportunism in the community selection of sites to establish as protected areas
1. Within Zonae Cogito, zoom in to community's area of tenure. Using a high-resolution satellite image as a backdrop for both the presentation of input data and solutions provides communities with the necessary spatial context. Within their area of tenure, landowners can quickly see priority sites (selection frequency) based on the current state of the protected areas network, and see what conservation features are contributing toward these priority areas (Figure 3).
2. If the community is flexible about the location of the protected area, then the best site can be chosen based on the current Marxan selection frequency. The scenario editing tool within Zonae Cogito can then be used to add this site to the protected areas network and recalculate target achievement for future negotiations.
3. Alternatively, if the community has a different site in mind for protection, perhaps because it is a traditionally protected tambu (sacred) site, or a resource they particularly want to conserve, for example, a forest with canoe trees (Gmelina molucana), then the Zonae Cogito scenario editor can also be used to evaluate immediately what this site contributes towards the protected areas network.
4. Equally, by using the same scenario editing method to lock sites out of the protected areas network, communities and the Lauru Land Conference of Tribal Communities can rapidly assess the consequence for the protected areas network of losing certain areas through logging, gardening, road construction, or other forms of development.
5. After agreeing on the site of a protected area, and including it in the network using the scenario editor, Marxan is rerun through Zonae Cogito, such that the assessment is dynamic and the next set of decisions (made sequentially) is informed by an up-to-date prioritization.
6. This process is then repeated as each new conservation opportunity arises.
Zonae Cogito, uses an integrated open-source geospatial information system (GIS), such that Marxan generated solutions can be rapidly viewed and edited without needing to interact with an independent GIS (Fig. 3). This functionality greatly reduces the technical expertise and systems needed to use the spatial prioritization software. Although housing the underlying data and building the Marxan databases currently relies upon The Nature Conservancy staff, successful execution of the process described in Table 1 depends upon capacity using Zonae Cogito being transferred to the Environment Officer of the Lauru Land Conference of Tribal Communities.
The reaction of Choiseul communities to the use of decision support software to help develop the protected area network has been overwhelmingly positive. Following initial work with Zonae Cogito to identify priority sites, comments from community members indicated they particularly appreciated the way the planning process and software was able to integrate their local knowledge together with external conservation values, on a single and equal platform. For many community members, the visualization element was integral to this appreciation; seeing all the information clearly displayed provided validation that their knowledge and interests were being represented and decisions were not preordained, and it greatly helped in building a belief that the creation of individual protected areas were leading to something larger. Rather than proving a hindrance, the technical aspect of working with computer software was perceived as adding to the gravitas with which parties were approaching the process.
Trials of the approach described in Table 1 in Choiseul revealed that the boundary length modifier (a variable used to control how much emphasis is placed on spatially cohesive protected area systems, see, Game & Grantham 2008), needed to be set close to zero (∼0.1). A low boundary length modifier helped improve the spread of priorities across communities.
There is a perception that the complexity and “black box” nature of spatial prioritization software can be an impediment to local engagement and decision making (Possingham et al. 2000). Rather than being a barrier to community-driven protected areas development, the ability of systematic approaches to rapidly identify both complementary and alternative conservation priorities, and to integrate community knowledge and values, provides a strong platform for collaborative conservation action.
The commitment by the communities of Choiseul both to establish a series of protected areas over the next 2 years, and do so in the systematic manner described here, will, in a few years time, provide an excellent opportunity to retrospectively evaluate the performance of informed opportunism as an approach to conservation planning. Important aspects of this evaluation will include the complementarity of protected areas established over the period, the capacity and enthusiasm of local staff to continue working with the software and the support needed for this to happen, and the communities’ satisfaction with, and continued support of, the site selection process.
Implementing a representative protected areas network through informed opportunism has a number of potential strengths and weaknesses. It does not necessarily guarantee that representation will be achieved. Allowing priorities to be widely dispersed across the planning area mean that initially (and at present in the case of Choiseul), there appears to be many equally good locations for protected areas. As network development proceeds, complementary decisions will become more restrictive, increasing the possibility that required sites will be unavailable. At the same time, there is a risk that early communities will only support protection of areas with low opportunity cost, making complete representation difficult. For this reason, it is equally important to update the assessment with areas that are “locked-out” of the network, as it is to include new areas being added.
On the other hand, because the consequences of all decisions, either for or against conservation, can be rapidly assessed, there is likely to be strong social pressure to make decisions that work toward target achievement at the whole island scale. Any unique features within a community's tenure will be evident, hopefully acting as a source of pride, rather than a conservation liability. Whether this is true will depend in part on the success of the environmental awareness campaign being conducted simultaneously.
The technical expertise required by end users of the Zonae Cogito decision support system is less than that required to formulate spatial prioritization problems and their related Marxan datasets. This distinction allows decision support systems to be widely used by decision makers with little technical expertise, hopefully maintaining the utility of conservation plans for supporting decisions. There is likely to be positive feedback in this regard, as the more communities, managers, and officials depend on such plans to help make decisions, the more impetus there is to update them. Providing the ongoing technical expertise required to develop and maintain the underlying data sets, while at the same time building local partners’ capacity to use accessible decision support tools such as Zonae Cogito, is an important role for conservation NGOs in countries such as the Solomon Islands.
Efficiency is one of the hallmarks of systematic conservation planning (Margules & Pressey 2000; Possingham et al. 2006), and the field's literature routinely focuses on the “efficiency” of protected area solutions delivered by particular planning approaches (e.g., Cameron et al. 2008; Leathwick et al. 2008; Ban et al. 2009; Lourival et al. 2009; Wiersma & Nudds 2009). A criticism of opportunistic protected area selection is that the end result is spatially inefficient, often requiring more area to reach the same target (Stewart et al. 2007). The use of optimization software to find spatially efficient protected area solutions is motivated by a belief that such solutions stand a greater chance of being implemented (e.g., Fernandes et al. 2005). In some countries, spatial efficiency will improve the social acceptability of solutions as a result of reduced opportunity and management costs. On the other hand, spatial efficiency is almost certainly a poor indicator of the social acceptability of a protected area network in the Solomon Islands and many other developing countries. No matter how spatially efficient a solution might be, if it does not get implemented, it is by definition inefficient.
However, a dispersed protected area network, likely to be the result of the process described here, also carries ecological risks. Small protected areas may be inadequate to sustain some species or communities (Brashares et al. 2001) and are more vulnerable to catastrophic disturbance events (Allison et al. 2003). Opportunistic support of small protected areas needs to be balanced by confidence that they can deliver conservation outcomes for the targeted species or communities. Initial monitoring results and community reports from the nine already established protected areas on Choiseul, suggest that they are providing effective protection, especially of harvested species such as Trochus shell (Trochus niloticus), and those sensitive to disturbance, such as megapodes.
The spatial prioritization approach developed here is neither a departure from established systematic planning principles, nor the technology used to support systematic conservation decisions. Rather, it is a departure from a static vision of conservation priority. Systematic conservation plans help decision makers to work transparently and efficiently toward conservation objectives. They should not, however, prevent decision makers from remaining responsive to conservation opportunities as they arise. We encourage more conservation plans to facilitate conservation decisions in a genuinely dynamic, interactive, and collaborative fashion.
The Nature Conservancy's work in the Solomon Islands would not be possible without the ongoing support of the Lauru Land Conference of Tribal Communities, the Solomon Islands National Government, and the Choiseul Provincial Government. This work and manuscript was improved by discussion with R. Pressey, and through comments from A. Knight, R. Noss, and an anonymous reviewer.