Coral-reef conservation planning in regions with high resource dependence: integrating lessons from socioeconomic and biodiversity approaches

Addressing the current biodiversity crisis is challenging. Recent advances in the field of conservation science have emphasized the need to consider whole social-ecological systems, recognising and accounting for the intricate relationships between nature and people. However, even with these well-justified calls for more multidisciplinary and complex approaches to conservation decision-making, the resources dedicated to conservation are still limited. Limited resources make cost-effective conservation essential, and this calls for planning. Conservation planning, the process of deciding where, when, and how to allocate limited resources to reduce biodiversity loss, is a burgeoning field with worldwide applications in a range of realms and contexts. The past decade saw a boom in systematic approaches to protected-area design, with decision-support tools proposing cost-effective sets of candidate reserves achieving explicit, quantitative conservation objectives (to allocate limited resources more wisely) for the least socioeconomic cost to affected stakeholders (to consider human needs). One of the foundations for the success of protected areas designed using a systematic approach is relevant, high-quality spatial datasets: on biodiversity to maximise its protection, and on the socioeconomic costs of implementing these protected areas to minimise potential negative impacts on resource-dependent human communities. Mimicking approaches used in the terrestrial and temperate marine realms, and to minimise the difficulty and cost of data collection, planners in coral-reef regions often use coral-reef habitats as proxies for marine biodiversity, and lost fishing opportunities as proxies for socioeconomic costs incurred by resource users. These applications are often based on a number of untested assumptions which, if false, could have serious implications for the actual success of protected areas. In coral-reef regions where most of the marine ecosystem is used by coastal communities, reconciling conflicting conservation and socioeconomic objectives is particularly challenging. Although the need for compromises on both sides is well known, the extent of trade-offs remains unclear. In Chapter 2, I developed a new method to quantify these compromises in Wallis, Alofi, and Futuna, three small Polynesian islands of varying geomorphologic and socioeconomic contexts, where threatened coral reefs are critical for subsistence fishing. Using Marxan, a reserve selection algorithm, and following current international habitat conservation guidelines, I designed protected-area systems to protect 20% of each coral-reef habitat type, progressively allowing an increasing proportion of fishing grounds to be reserved. Stronger trade-offs between conservation and socioeconomic objectives appeared when reserves were designed to protect habitats mapped with more detail, and using larger management units. The extent of trade-offs also varied according to local socioeconomic and geomorphologic contexts in the three islands. By demonstrating that the ability to achieve conservation and socioeconomic objectives and extent of compromises needed are largely data and context-dependent, I highlighted the need to question the effect on and relevance to objectives of different biological and socioeconomic data used in local-scale coral-reef conservation planning. In Chapters 4, 5, and 6, I investigated the implications of using different socioeconomic and biological data based on common assumptions about local-scale coral-reef conservation decisions. I used the Madang Lagoon in Papua New Guinea as a case study because 1) this area, covering 40 km2, is situated in the Coral Triangle (a known hotspot for coral-reef biodiversity), 2) local coastal communities maintain close relationships with their marine environment, and 3) a large international biodiversity expedition in the region facilitated logistics for my data collection, as well as access to a comprehensive set of data rarely available in such contexts. I detailed data collection methods in Chapter 3. To minimise impacts on fishers, protected areas are often placed away from the most important fishing grounds. A main assumption is that placing important fishing places for fishers inside protected areas will likely incur high socioeconomic costs. These costs are typically quantified through proxies of current fishing opportunities such as proximity to villages, fishing effort, total catch, or average catch per unit effort. In Chapter 4, I investigated the validity of such proxies as indicators of the importance of fishing places for fishers, thus of potential socioeconomic costs if placed inside a protected area. I mapped fishing grounds and their relative importance defined 1) according to the proxies, and 2) as perceived by fishers themselves, and compared them. I found strong spatial differences between reserves designed to minimise costs based on proxies and the perceptions of fishers. There are two possible explanations, both of which have serious implications for planning. First, fishers' perceptions could incorporate spatiotemporal variations in fisheries patterns that are hard to capture with short-term data collection (commonly done for conservation planning and in this study). Second, each dataset tells planners different things, but both could be valid, meaning that both perspectives will be relevant to different objectives. For instance, proxies based on catch data could help planners achieve objectives related to economics and food security by focusing on maintaining fishing effort and the quantity of resource extracted, while fishers' perceptions could help achieve more social objectives by including fishers' values and preferences. People access and value their marine environment for more than just fishing, especially in coastal coral-reef areas. For example, coral reefs are also accessed for recreation, spirituality, or their aesthetic value, and harvested for cultural purposes. In Chapter 5, I tested the adequacy of using only the importance of places for fishing as a socioeconomic cost of reserve implementation. My hypothesis was that implementing protected areas not only incurs costs to fishers through constraints on fishing, but also to the broader community through revoking harvest and/or access to places that provide other ecosystem services like recreation, traditional medicine, spirituality. I developed a new approach to map and quantify the perceived importance of different places for community members according to all these benefits, including fishing. Similar approaches exist for extensive terrestrial regions in developed countries, promoting the use of such data in conservation planning. However, none of these approaches explicitly demonstrates how this information can be incorporated into planning. I developed a novel method to do so, and demonstrated that locating reserves to minimise costs to fishers is likely to incur significant costs to the wider community by displacing reserves into areas where access to other benefits is important. This result has major implications, since I am demonstrating that this common approach can provide a false sense of achievement of socioeconomic objectives. Without including information on all benefits derived from coral reefs, there is a clear risk of compromising values of local people and undermining the cooperation needed for reserves to be effective in biodiversity conservation. In conservation planning, it is often assumed that designing reserve systems that encompass a greater diversity of habitats will incidentally protect a higher number of species. It is also assumed that finer-resolution and more complex data are more representative of "the truth", implying that representation of "true" biodiversity and reduction of "true" socioeconomic cost increases with effort in data collection. In Chapter 6, I tested these two assumptions. To do this, I compiled the most comprehensive empirical dataset available for a single small coral-reef region, in the Madang Lagoon, in Papua New Guinea. To use as biodiversity proxies, I created different habitat maps for the region using satellite imagery and ground-truth data. As a reference biodiversity dataset, I compiled georeferenced species lists for macro-algae, corals, fish, and macro-invertebrates collected by taxonomists in the region. As proxies for socioeconomic costs, I compiled the different datasets developed in Chapter 4. As a reference socioeconomic cost, I used the perceived importance of places for derived ecosystem benefits developed in Chapter 5 because these data gave the most comprehensive picture of reef uses and values to local people. Using only areas for which I had all four datasets, I designed reserve systems that aimed at protecting biodiversity proxies while minimising socioeconomic cost proxies, using all possible combinations of data types. I created the Proxy Effectiveness Index (PEI) to measure the effectiveness of such reserves at protecting reference biodiversity and at reducing reference socioeconomic cost. For biodiversity, I found that using more detailed habitat maps does not necessarily lead to representing more species in candidate reserves and that surrogacy effectiveness was highly variable, suggesting the interaction of numerous other factors. Reducing socioeconomic costs based on catch data performed best to reduce the reference cost compared to coarser proxies. Finally, the most expensive combinations of biodiversity and cost proxies do not necessarily provide the most cost-effective reserve systems (i.e. best reference biodiversity representation for the lowest reference socioeconomic cost). I emphasised that obtaining a robust measure of the cost-effectiveness of reserves based on proxies is difficult because of the global lack of reference datasets, and results will vary with numerous factors inherent to the data used for testing and to the test method itself. My thesis has a number of implications for biodiversity, conservation planners, and human coastal communities in tropical regions, which I discuss in Chapter 7. The implications arise from the risk that identifying candidate protected areas using inadequate data can lead to a false sense of achievement of both conservation and socioeconomic objectives, ineffectively protecting biodiversity while incurring significant impacts on local communities. First, tropical fisheries are complex and highly variable in space and time, and connections between people and coral reefs go well beyond fishing, highlighting the need for socioeconomic assessments that more comprehensively and accurately reflect the perceptions and values of local communities. Second, planners must carefully consider the social, economic, and even cultural goals of their reserves to influence their choice of proxies for socioeconomic costs. Third, protecting habitats may be a good way to achieve biodiversity protection, but the effectiveness of habitats as surrogates of biodiversity cannot be assumed unless it is tested in a similar geomorphologic and socioeconomic context. Finally, return on investment in data collection does not necessarily increase with the cost of data, so planners should focus on relevance, rather than quantity or level of detail, in gathering important information for conservation