Standardizing the evaluation of community-based conservation success

Abstract

Community-based conservation, which strives to simultaneously improve nature conservation and alleviate poverty, must provide biological and socioeconomic benefits that are linked through effective resilience mechanisms. To date, few community-based conservation initiatives have published comprehensive assessments that track performance in these elements of success. With 45% of the world's protected areas in comanagement with local communities, standardized measures to effectively evaluate the dual goals of community-based conservation are needed. We here introduce SPECCS, a user-friendly Standardized Protocol for Evaluating Community Conservation Success that incorporates an appraisal of data quality to responsibly assess progress over time or to compare effectiveness among different initiatives. We illustrate SPECCS's use by evaluating the Wechiau Community Hippo Sanctuary (WCHS) of northern Ghana 10 and 20 yr after its inception. The WCHS has the dual objective of protecting one of Ghana's few remaining hippopotamus populations while alleviating poverty in the surrounding communities through the creation of economic opportunity and infrastructure development. Results suggest stable project performance in the 10-yr (76%) and 20-yr (76%) evaluation, with an improvement in evaluation quality from 30% to 34%. The project is currently stronger in socioeconomic (performance 86%; quality 30%) than biological (60%; 32%) outcomes and in benefits (83%, 42%) than resilience (63%, 21%). Biological resilience is challenged by poor connectivity and limited project control over threats, whereas socioeconomic resilience is affected by a decision balance that continues to favor external stakeholders. SPECCS helps pinpoint strengths and weaknesses for timely adaptive management, strategic investments, and evidence-based recognition of community-based conservation successes.

Introduction

Modern-day conservation initiatives often aim not only to safeguard endangered species and habitats, but also to accommodate people's needs. This reflects a paradigm shift away from stringent protectionism following recognition in the 1980s of the importance of local support in fostering conservation success, and concerns over the burden that conservation can impose on the world's poor, especially given widespread spatial overlap between poverty and threats to biodiversity (Sachs et al. 2009, Miller et al. 2011, Gurney et al. 2014). The resulting integrated approach, here referred to as community-based conservation, takes many names and guises, but fundamentally aims to simultaneously protect nature and alleviate poverty. Although the approach has long been criticized for failing people, biodiversity, or both (Wells and McShane 2004, Miller et al. 2011), recent systematic reviews have found more evidence for success than failure (Brooks et al. 2013, Oldekop et al. 2016, Brooks 2017).

Win-win scenarios for both people and nature appear to be fostered by institutional capacity building, participatory approaches, and environmental education, but also project age (Brooks 2017). This highlights the potentially dynamic nature of trade-offs between ecological and socioeconomic goals. Such dynamic trade-offs demand strategic monitoring of impacts on nature and people to allow for timely adaptive management (Plummer and Armitage 2007, Sheppard et al. 2010, Brooks et al. 2013, Brooks 2017). Unfortunately, performance evaluations of community-based conservation at multiple time points are rare, as are comprehensive assessments that evaluate ecological and socioeconomic outcomes (Brooks et al. 2013, Gurney et al. 2014, Oldekop et al. 2016, Brooks 2017).

Previous performance evaluations assessed whether initiatives achieved their stated objectives (Brooks et al. 2013, Oldekop et al. 2016, Brooks 2017). Given large variation in the type and number of objectives (Brooks et al. 2013), a standardized approach for appraising performance would be helpful in comparing success over time or among projects. Standardized evaluation seems especially critical given the many protected areas worldwide that incorporate sustainable resource usage and comanagement by local communities. Of the 34,700,000 km² covered by the world's protected areas as of 2016, 45% are classified as category V (7,300,000 km² or 21%) or category VI (8,300,000 km² or 24%); these likely incorporate both biological and socioeconomic goals (UNEP-WCMC and IUCN 2016).

We propose a Standardized Protocol for Evaluating Community Conservation Success (SPECCS) as a user-friendly evaluation tool that guides practitioners and researchers through the complexities of a comprehensive performance assessment. SPECCS yields standardized performance scores alongside an indicator of evaluation quality. It is designed to accommodate both variable data availability and the diversity of project-specific goals that pursued by community-based conservation initiatives. Evaluation outputs allow projects to track progress over time, highlight areas of strength and weakness for adaptive management, and facilitate comparisons between initiatives to advance research regarding factors that influence success. To illustrate SPECCS's utility, we employ it in a longitudinal study that compares the performance of the Wechiau Community Hippo Sanctuary (WCHS) in Ghana 10 and 20 years after inception. Dual objectives of WCHS are to protect the local population of hippopotamus (Hippopotamus amphibious) and to alleviate poverty in the surrounding communities via a landscape approach that integrates a strictly protected core zone with infrastructure developments that yield sustainable economic opportunities for residents in an adjacent development zone.

SPECCS is a web-accessible evaluation tool that is novel in addressing data quality while integrating many of the criteria identified as important components of community conservation success in previous literature (Walpole and Wilder 2008, Waylen et al. 2010, Gurney et al. 2014, Baynes et al. 2015, Brooks 2017, Reid et al. 2017; A. Fishman, G. Haber, F. Hervas, E. Winata, unpublished manuscript) or deemed valuable by the United Nations Equator Prize and Darwin Initiatives (Equator Initiative 2017, UK Government 2017; A. Fishman, G. Haber, F. Hervas, E. Winata, unpublished manuscript). SPECCS touches on all 17 United Nations Sustainable Development Goals (United Nations 2015) and reflects all five of the Aichi Targets’ strategic goals (Convention on Biological Diversity 2017). Its evaluation criteria thus address common threats to human well-being and biodiversity (Fig. 1) and are grouped into five elements: biological benefits, biological resilience, socioeconomic benefits, socioeconomic resilience, and linkage mechanisms (Walpole and Wilder 2008, Sheppard et al. 2010). Assessment of multiple criteria per element, and flexibility regarding specific evaluation indicators, ensure both comprehensive assessment and project-specific relevance. Fully recognizing the constraints on funding, scope and monitoring capacity under which many community conservation initiatives operate (Brooks et al. 2013), SPECCS is applicable in both data-poor and data-rich situations.

Figure 1

Common threats to human well-being and biodiversity, relevant mitigating factors, and their relationship to the evaluation criteria used by the Standardized Protocol for Evaluating Community Conservation Success (SPECCS).

Methods

SPECCS

SPECCS is available online as a user-friendly web tool or as an interactive spreadsheet (available online).1 The tool guides users through scoring each of 23 performance criteria on an ordinal scale from 0 to 3 (Table 1). Simultaneously, the quality of data used to evaluate each criterion is assessed by multiplying an indicator of analytical quality (no data, 0; anecdotal data, 1; qualitative data or bivariate analysis, 2; and multivariate quantitative or mixed methods analysis, 3) with an indicator of study design (no data, 0; no control, 1; spatial or temporal control, 2; spatial and temporal control, 3). Performance and quality scores for each criterion, respectively, are then tallied and expressed as a percentage of theoretically possible totals to calculate overall and element-specific scores. Performance totals are scored based only on evaluated criteria and hence not penalized if data for a particular criterion are unavailable. Fair comparisons between data-rich and data-poor projects are possible given the associated quality score.

Table 1. An overview of evaluation criteria and how to score them

Criterion	Definition	Score 0	Score 1	Score 2	Score 3
Biological benefits
1. Trend in target of protection	What is the trend in the specific taxon or habitat targeted for conservation by the project in terms of population measure or extent?	Declining as severely as or worse than at project initiation or last evaluation.	Declining but less severely as at project initiation or last evaluation.	Stable but likely below historic maxima or carrying capacity.	Increasing since project initiation or last evaluation OR Stable at historic maxima or carrying capacity
2. Trend in threats to target of protection	What is the trend in threats to the target of protection?	Pre-existing threats are worsening and/or additional threats have arisen since project initiation or last evaluation.	There are no new threats and existing threats are at similar intensity as at project initiation or last evaluation, or some threats have been mitigated but others have emerged or become stronger resulting in a stable level of exposure to threats.	Threats remain but the number of threats and/or their combined intensity is less than at project initiation or last evaluation.	All threats have been eliminated
3. Habitat quality for taxon targeted for protection or of the habitat targeted for protection	How degraded is the habitat and what is the trend in its quality, as indicated, e.g., by percentage native vegetation cover, indices of biotic integrity, hydromorphology, live coral cover, benthic rugosity, degree of human modification, percentage of invasive species?	More than 50% of the habitat is degraded and not improving.	<50% but more than 30% of the habitat is degraded and not improving or >50% of the habitat is degraded but improving.	30–50% of the habitat is degraded but improving or <30% of the habitat is degraded but deteriorating.	<30% of the habitat is degraded and stable or improving
4. Umbrella benefits to biodiversity or natural capital more widely	Are there any benefits to nature beyond the specific target of protection?	Harmful effects on nontarget native species, habitat, or natural capital causing conservation concerns.	No benefits to nature beyond the specific target of protection but no harmful effects on nontarget native species, habitat, or natural capital causing conservation concerns.	Weak or few beneficial effects to nontarget native species, habitat, or natural capital.	Strong or many benefits to nontarget native species, habitat, or natural capital
Biological resilience
5. Connectivity/viability of the target of protection	What is the connectivity or viability of the target species or habitat?	The target of protection is isolated or highly fragmented and its viability within the project area is questionable on grounds such as gender distribution, genetic composition, or small population size.	The target of protection is isolated or highly fragmented but its viability within the project area is assured in the short term on other grounds such as sufficient population size and adequate genetic composition or the target of protection has some connectivity but its viability within the project area is threated on grounds such as gender distribution, genetic composition, or small population size.	The target of protection has some connectivity, and additionally its viability within the project area is assured in the short to intermediate term on other grounds such as sufficient population size and adequate genetic composition.	The target of protection is highly connected or continuous over a sufficiently large area and its viability within the project area is assured in the intermediate to long-term on other grounds such as sufficient population size and adequate genetic composition
6. Structural integrity of the species or habitat targeted for protection	(a) If the target of protection is a specific taxon, how diverse is the age, size or life stage structure of the population? (b) If the target of protection is a habitat, how diverse are the habitat's foundation and/or key-stone species in age or taxonomic or functional composition?	(a) The entire population falls within one age group, size or life stage. (b) Critical foundation species, ecosystem engineers or key-stone species have minimal age diversity (as defined above) or minimal (<40%) taxonomic or functional diversity as compared to the composition of a pristine version of the target habitat.	(a) The population includes several but not all age groups, sizes and life stages, leading to an imbalanced structure as compared to a healthy population (e.g., intermittent reproduction). (b) Critical foundation species, ecosystem engineers or key-stone species have imbalanced age diversity (as defined above) or comprise 40–70% of taxa/functional groups that would occur in a pristine version of the habitat, but lack specific rare, vulnerable, or functionally unique taxa.	(a) The population includes all age groups, sizes or life stages, but some are likely underrepresented relative to a healthy population's structure. (b) Critical foundation species, ecosystem engineers or key-stone species benefit from the presence of all age groups but in atypical proportion (as described above) or comprise 40–70% of taxa/functional groups that would occur in a pristine version of the habitat, including specific rare, vulnerable, or functionally unique taxa.	(a) The population includes all age groups, sizes or life stages and likely is appropriately distributed among them (e.g., consistently sufficient reproduction) (b) Critical foundation species, ecosystem engineers or keystone species benefit from a healthy age structure or comprise >70% of taxa/functional groups that would occur in a pristine version of the habitat, including specific rare, vulnerable, or functionally unique taxa, functional groups, or life forms.
7. Project control over threats	How much control or influence over the protection of the target species or habitat does the project entail?	The project addresses <10% of all potential threats to the target of protection or reduces their combined severity by <10%.	The project addresses 10–50% of all potential threats to the target of protection or reduces their combined severity by <50%.	The project addresses 50–90% of all potential threats to the target of protection or reduces their combined severity by >50%.	The project addresses the vast majority (>90%) of all potential threats to the target of protection, or dramatically (>90%) reduces their combined severity.
Socioeconomic benefits
8. Access to amenities	Has access to amenities, such as clean water, healthcare, schools, etc., improved since project initiation or last evaluation?	Access to amenities is the same as at project initiation or last evaluation or has become worse.	Access to amenities has improved for a few types of amenities (e.g., only education, not sanitation, health care, etc.) and for only few people (<35%).	Access to amenities has improved for a few types of amenities (e.g., education and market access but not health) and <70% of people.	Access to amenities has improved in multiple areas for most people (>70%).
9. Employment creation	Have any new employment or income-generating opportunities been created as a result of the project since its initiation or last evaluation?	No new income-generating opportunities were created and/or certain occupations were lost without replacement as a result of the project.	The project has created a few income-generating opportunities for the community but the number of opportunities created equates to <10% of the project area's population size.	The project has created several income-generating opportunities for the community but the number of opportunities created equates to <25% of the project area's population size.	Many new income-generating opportunities have been created for members of the community as a result of the project with opportunities available for 25% or more of the project area's population.
10. Capacity building	Since its initiation or last evaluation, has the project increased capacity among members of the community, e.g., via education, training, provision of advanced tools or technology, establishing decision-making or problem-solving processes?	No capacity building has occurred.	Capacity building has occurred but benefits are centered on the select individuals that have received training/education/tools and are not reaching the wider project community.	Widespread capacity building has occurred but benefits are centered only on the individuals that have received training/education/tools without community-wide benefits.	Widespread capacity building with both individual benefits for most and community-wide benefits.
11. Fostering cultural diversity	Has the project acknowledged, respected, and supported unique cultural diversity within the area since its initiation or last evaluation?	Imposition of external values in disregard of local and/or traditional norms, beliefs, and knowledge.	Some but not all aspects of the project are guided by the local and/or traditional norms, beliefs, and knowledge of some but not all locally represented cultures or insincere promotion of local culture for commercial gain in absence of substantive integration in project design.	Most but not all aspects of the project are guided by or compatible with the local and/or traditional norms, beliefs, and knowledge of all locally represented cultures or all aspects of the project are guided by or compatible with the local and/or traditional norms, beliefs, and knowledge of some but not all locally represented cultures.	All aspects of the project are guided by or compatible with the local and/or traditional norms, beliefs, and knowledge of all locally represented cultures.
Socioeconomic resilience
12. Community rights to land and resources	How empowered is the local community in terms of legal recognition of land title and resource management powers?	Absence of community empowerment, with neither de jure nor de facto recognition of land tenure and/or resource management decision-making powers.	Community empowerment on paper, with de jure but not de facto rights to land tenure and/or resource management decision-making powers.	Community empowerment but long-term uncertainty, with de facto but not de jure land tenure and/or resource decision-making powers.	Strong local or community empowerment, with de jure and de facto land tenure and resource management decision-making powers.
13. Capacity to enforce project rules and regulations	Does the community have the necessary legal, financial, logistic, and human resources to enforce the rules and regulations required to achieve biological and socioeconomic project goals?	The community has no legal, financial, logistic, and human resources to enforce the rules and regulations required to achieve biological and socioeconomic project goals.	The community has some but, in most cases, insufficient legal, financial, logistic, and/or human resources to enforce the rules and regulations required to achieve biological and socioeconomic project goals.	The community often but not always has the necessary legal, financial, logistic, and human resources to enforce the rules and regulations required to achieve biological and socioeconomic project goals.	The community consistently has the necessary legal, financial, logistic, and human resources to enforce the rules and regulations required to achieve biological and socioeconomic project goals
14. Decision balance between the local community and external stakeholders	Who has the power to make decisions about the direction and development of the project?	External stakeholders are the sole decision makers regarding project development and/or future directions.	Both external stakeholders and local leaders or the local community contribute to decisions about project development and/or future directions but external stakeholders are the primary decision makers.	Both external stakeholders and local leaders or the local community contribute about equally to decisions about project development and/or future directions.	Local communities are the primary or sole decision makers regarding project development and/or future directions.
15. Equitability of benefit sharing	How are the benefits of the project distributed across the project area's population?	Benefits reach only a select, already advantaged sub-group of the local community (e.g., the dominant ethnic group, gender, age group, or geographic area within the community as a whole).	Benefits reach only select sub-groups of the local community but not necessarily the most advantaged (e.g., in terms of ethnic group, gender, age, geographic distribution within the community as a whole).	Benefits reach many sub-groups of the local community but not the traditionally disadvantaged (e.g., in terms of ethnic group, gender, age, geographic distribution within the community as a whole).	Benefits reach all sub-groups of the local community including traditionally disadvantaged groups (e.g., in terms of ethnic group, gender, age, geographic distribution within the community as a whole).
16. Economic sustainability	How dependent is the project on external financial support in the short and long-term?	The project is completely dependent on external funding support with 100% of its financial needs being met externally and no prospect for financial self-sufficiency within the next three years.	The project can finance less than one-half (<50%) of its needs without external funding support and/or financial self-sufficiency is dependent on a short-lived source of income likely to become insufficient within the next three years.	The project can finance more than one-half (>50%) of its needs and/or self-sufficiency is likely to be achieved within the next three years.	The project is financially self-sustaining, generating enough funds to cover project expenses and ideally a surplus, and will remain so for the next three years or more.
17. Income diversification	Is the project's funding dependent on a single source or set of sources all sensitive to the same market forces?	All project-generated income or external funding support comes from a single source or set of sources all sensitive to the same market forces (e.g., multiple ecotourism-dependent ventures).	Project income and/or external funding comes from more than one source or set of sources governed by distinct market forces, but the sources are inconsistent.	Project income and/or external funding derive from several consistent sources or sets of sources but with considerable overlap in their sensitivity to market forces.	Project income and/or external funding derive from more than one consistently profitable source or set of sources governed by distinct market forces (e.g., an ecotourism venture and an export product).
18. Social capital to problem solve	Is there sufficient bonding and bridging social capital to effectively explore, acknowledge, and mitigate vulnerabilities and how well are challenges resolved when they occur?	No acknowledgement of factors that could destabilize the project and no or inappropriate response when challenges arise.	Awareness and some preventative action with regard to some but not all potential threats to the project frequently resulting in a failure to minimize damage.	Awareness and some preventative action with regard to many potential threats to the project but repeated occurrence of responses insufficient to minimize damage.	Awareness and preventative actions with regard to a comprehensive list of potential threats to the project, with appropriate, timely, and sufficient actions taken to minimize damage.
Linkage mechanisms
19. Ecological awareness	How knowledgeable are local people or communities about their natural environment and the importance of conservation?	The local community have no understanding of the surrounding environment and/or is unfamiliar with or opposed to the idea of protecting imperiled species and/or habitats.	The local community has some but no thorough understanding of the surrounding environment and/or is open to but not actively embracing the idea of protecting imperiled species and/or habitats.	The local community has some but no thorough understanding of the surrounding environment but explicitly recognizes the importance of actively protecting imperiled species and/or habitats.	The local community is knowledgeable about the surrounding environment and explicitly recognizes the importance of actively protecting imperiled species and/or habitats.
20. Attitudes and emotional investment	How do local communities feel about the project and are they emotionally invested in the project's success?	Most community members (>75%) dislike or feel resentment toward the project and most have no interest in or are opposed to the project reaching its goals.	There are mixed feelings among community members about the project, with 25–50% of people interested and/or taking pride in the project reaching its goals.	There are mostly positive attitudes among community members about the project, with 50–75% of people interested and/or taking pride in the project reaching its goals.	Most community members (>75%) have a positive attitude toward the project and are interested and/or take pride in the project reaching its goals.
21. Tangible linkage between biological and socioeconomic outcomes	How interdependent are biological and socioeconomic outcomes of the project?	Biological benefits and socioeconomic benefits are not interdependent, with the possibility that one type of benefit could succeed while the other fails.	A minority of biological and socioeconomic benefits are linked artificially (e.g., because external technical or financial support for socioeconomic benefits has been made contingent on reaching or maintaining specific biological outcomes), leaving the risk that benefits that are not linked may cause an imbalance of progress for either biological or socioeconomic benefits.	All biological and socioeconomic benefits are linked artificially, leaving the risk that leniency on artificial links may cause an imbalance of progress for either biological or socioeconomic benefits or a minority of biological and socioeconomic benefits are linked physically, leaving the risk that benefits that are not linked or weakly linked artificially may cause an imbalance of progress for either biological or socioeconomic benefits.	The majority of biological and socioeconomic benefits are physically linked, with one unable to succeed without the other, e.g., because income-generation is dependent on a thriving, healthy environment and successful protection of the target taxon or habitat (e.g., eco-tourism, carbon-credit payments, etc.).
22. Local perception of the interdependence between biological and socioeconomic outcomes	Do the local communities perceive that socioeconomic and biological benefits are interdependent?	Most community members (>75%) do not perceive any linkage between socioeconomic and biological benefits.	Some community members (25–50%) perceive a link between socioeconomic and biological benefits, but these links are mostly indirect (e.g., recognizing that support by external stakeholders reflects conservation interests).	Many community members (50–75%) perceive a link between socioeconomic and biological benefits, including physical links (e.g., recognizing that food security is dependent on successful protection of the target taxon or habitat).	The majority of community members (>75%) perceive a direct physical link between socioeconomic and biological benefits, recognizing, for example, that food security is dependent on successful protection of the target taxon or habitat.
23. Investment of project-derived socioeconomic gains in conservation	Are the socioeconomic gains (monetary profit, time, or knowledge gains) derived from the project invested in conservation, for example by increasing protective measures or the area under protection?	Socioeconomic gains derived from the project are not being invested in the maintenance or improvement of conservation.	Few (<20%) socioeconomic gains derived from the project are being invested in the maintenance or improvement of conservation.	Several (20–50%) socioeconomic gains derived from the project are being invested in the maintenance or improvement of conservation.	50% or more of socioeconomic gains invested in the maintenance or improvement of conservation.

Note

• Criteria are grouped into five elements of success: biological benefits, biological resilience, socioeconomic benefits, socioeconomic resilience, linkage mechanisms.

The choice of a four-point scale for evaluating performance arose as a compromise, given the diverse nature of criteria assessed. With criteria spanning multiple facets of biological benefits, biological resilience, socioeconomic benefits, socioeconomic resilience, and linkage mechanisms, a three-point scale along the lines of “worse, stable, better” seemed too crude to capture important nuances, whereas a five-point scale identifying “worst, bad, stable, good, and best” scenarios proved difficult to define in many cases. We therefore settled on defining four scenarios for each criterion, where 0 signifies situations that are deteriorating, not improving, or simply bad, 1 indicates situations that are stable, marginally improving or inadequate, 2 reflects good situations or major improvements and 3 represents the ideal case (Table 1).

SPECCS comprises four criteria evaluating biological benefits, i.e., direct improvements for the species or habitat targeted for protection of for nature generally (Table 1). Because the default definition for recovery is often an increase in population size or geographic distribution (Akçakaya et al. 2018), our first criterion examines trends in the target of protection, akin to criteria considered by Walpole and Wilder (2008) and Brooks (2017). Possible indicators include change in population numbers or more complex measures like the IUCN Green List (Akçakaya et al. 2018). Because improved conservation status and reduced conservation dependence are associated with accurate identification and removal of threats (Crees et al. 2016), our second criterion focuses on trends in threats (see related criterion of threat severity in Brooks 2017). Threats could include unsustainable use of natural resources, habitat loss or disturbance resulting from infrastructure developments, climate change impacts, invasive species, disease, or inbreeding depression. Indicators may be as simple as assessing change in the number of known risks to the target of protection, or as complex as a comprehensive risk assessment considering both likelihood and significance. With habitat degradation predominant among threats to biodiversity (Mortelliti et al. 2010, Chaudhary et al. 2018), our third criterion examines habitat quality (see related criteria in Gurney et al. 2014 and Brooks 2017). Possible indicators include percent native cover, percentage of invasive species or indices of biotic integrity, hydromorphology, rugosity, and degree of human modification. We set our upper threshold defining ideal habitat quality at 70% intact, based on Conservation International's definition of wilderness (Mittermeier et al. 2003). Lower thresholds additionally consider trends in quality to consider potential impacts of improvement or deterioration. Our final criterion for biological benefits considers umbrella benefits in line with the umbrella species concept (Branton and Richardson 2011) and resilience thinking (Curtin and Parker 2014) to reflect that efforts to conserve one species or habitat can have positive or negative implications for others. Possible indicators include trends in the abundance of nontarget native species, species richness, functional diversity, or ecosystem services (Brooks et al. 2006, Reid et al. 2017).

SPECCS also includes three criteria evaluating biological resilience to assess an initiatives’ capacity to perpetuate biological benefits in the long term (Curtin and Parker 2014; Table 1). The first considers both connectivity and viability of the target of protection. Low contiguity between habitat patches caused by the fragmentation of ecological landscapes counts among the greatest impediments to biodiversity conservation (Mortelliti et al. 2010, Gao et al. 2017), hindering dispersal, meta-population dynamics, and gene flow. Possible indicators include measures of patch size or isolation, such as distance to the next nearest population of conspecifics or landscape genetic analyses. Even where connectivity is good, long-term population viability, which reflects a populations’ ability to persist for one or more decades or multiple generations (Boyce 1992), may be poor. Possible viability indicators include measures of gender distribution, (effective) population size, genetic composition, or formal population viability analyses. Among previous authors, only Brooks (2017) touched on this criterion by considering increased habitat extent among indicators of ecological success. Our second criterion focuses on the structural integrity of the target of protection. For initiatives targeting focal species, structural integrity refers to the target species’ population composition in terms of age, size, or life stage. Age-dependent survival and recruitment rates are critical in regulating population dynamics and extinction risk (Gaillard et al. 1998). Similarly, the persistence of habitats can hinge on the age structure of essential foundation species or ecosystem engineers, with gaps or imbalances in age cohorts often indicative of previous or ongoing harvest or other human disturbances (Ellison et al. 2005). Where habitats targeted for protection depend on a multitude of foundation species, their taxonomic or functional diversity may represent an alternative indicator of structural integrity (Riemann et al. 2017). For these alternative indicators, we defined the ideal scenario as comprising >70% of the taxonomic or functional diversity of a pristine version of the habitat to mirror Conservation International's definition of wilderness (Mittermeier et al. 2003). To our knowledge, no previous authors considered structural integrity among success criteria for community conservation success. Our third criterion for biological resilience evaluates project control over threats to reflect the reality that even impeccable natural resource management at local scales may not suffice to safeguard species populations or habitats afflicted by regional or global factors such as pollution or climate change, necessitating consideration of broader-scale barriers (Brooks et al. 2012, 2013, Reid et al. 2017).

SPECCS incorporates four criteria evaluating socioeconomic benefits representing both tangible and intangible improvements to the local quality of life (Table 1). Our first criterion assesses access to amenities such as clean water, sanitation, schools, hospitals, and transport, communications, and energy infrastructure (see related criteria in Walpole and Wilder 2008, Gurney et al. 2014, Brooks 2017). Improved access to amenities is critical to ending poverty (The World Bank 2001). Our second criterion focuses on employment creation. Several previous authors have noted the importance of employment, income, and financial assets in community conservation success (Walpole and Wilder 2008, Brooks et al. 2013, Brooks 2017). We defined an ideal scenario as project-derived employment for 25% or more of the population, recognizing that preproject livelihoods need not all be replaced, that global rates of extreme and moderate working poverty sum to ~25%, and that regional unemployment rates rarely exceed 25% (ILO 2018). Our third criterion considers capacity building via transfer of knowledge, skills, or technology. The importance of individual skills development (Salafsky et al. 2001, Tallis et al. 2008) and strengthened institutional capacity (Agrawal and Benson 2011, Persha et al. 2011) is widely recognized (Campbell et al. 2010, Brooks 2017). It provides immediate benefits to people, improves the effectiveness of natural resource management, and motivates local participation (Baynes et al. 2015). In an ideal scenario, the majority of community members benefit directly or indirectly in terms of new skills or tools and ensuing community-wide improvements such as reduced conflict, increased economic activity, and better government liaison (Baynes et al. 2015). Our fourth criterion, fostering cultural diversity, acknowledges the importance of embracing cultural heterogeneity by respecting and integrating the local and traditional ecological knowledge, beliefs, and norms of all ethnic, faith, and other cultural groupings represented within the community. Dignified treatment of diverse cultural perspectives reinforces their value, instils local pride, promotes a sense of self-worth and ownership, and can aid the intergenerational transfer of traditional knowledge (Walsh et al. 2013, McPherson et al. 2016). Ignoring the local cultural context or exploiting sociocultural divisions to advance a particular agenda is therefore detrimental to effective community engagement (Mishra et al. 2017), and previous authors have considered appropriate integration of local cultural aspects key to community conservation success (Waylen et al. 2010, Brooks et al. 2012, 2013, Brooks 2017, Reid et al. 2017).

SPECCS uses seven criteria to evaluate socioeconomic resilience or an initiative's ability to persist and yield benefits over the long-term (Table 1). The first examines land and resource management rights. Security of tenure over land and resources lower the opportunity cost of sustainable resource management and have been considered among critical factors of community conservation success (Noss 1997, Waylen et al. 2010, Brooks et al. 2012, 2013, Gurney et al. 2014, Baynes et al. 2015, Brooks 2017). The second criterion assesses the capacity to enforce rules and regulations given governance mechanisms, government support, institutional, legal, financial, logistic, and human resources. Strengthened, context-appropriate compliance mechanisms are seen as important to community conservation success (Cinner et al. 2012, Gurney et al. 2014). Our third criterion helps address context relevance by assessing the decision balance between local and external stakeholders. Multiple studies point to the importance of participatory approaches that involve the local community in project initiation, design, and daily management (Waylen et al. 2010, Persha et al. 2011, Brooks et al. 2012, Brooks 2017). The fourth criterion, equitability of benefit sharing, is thought to foster social cohesion and social capital by reducing social angst and existing inequalities, protecting vulnerable groups, and enhancing participation (Waylen et al. 2010, Baynes et al. 2015, Brooks 2017, Reid et al. 2017). While we acknowledge that it may be important to consider cultural context in benefit distribution, we nonetheless believe that ideally initiatives find mechanisms to reward all members of the community equitably. Our fifth criterion, economic sustainability, acknowledges that many community conservation initiatives whither once short-lived external funding ends (Gurney et al. 2014), unless they have managed to secure a stable local source of income sufficient to sustain project management costs (Reid et al. 2017). The sixth criterion considers income diversification to minimize vulnerability to market volatility and thus increase socioeconomic resilience (The World Bank 2001, Gurney et al. 2014). The seventh criterion evaluates social capital to problem solve, including the ability to recognize, assess and effectively mitigate threats to the success of the community conservation initiative. This depends on a community's ability to jointly pursue common goals as well as connect with external players (bonding and bridging social capital, respectively (Baynes et al. 2015) and shapes their capacity to adapt (Reid et al. 2017)).

SPECCS's final five criteria evaluate linkage mechanisms that foster continued integration of development and conservation (Sheppard et al. 2010, Table 1). The first of these criteria focuses on ecological knowledge and awareness of the need for conservation, with environmental education and knowledge commonly considered key to success (Waylen et al. 2010, Gurney et al. 2014, Brooks 2017). Our second linkage criterion assesses attitudes and emotional investment, because public support is mission-critical in many ventures and can help tie the dual goals of community conservation together when feelings of pride and empowerment generate community buy-in (Sheppard et al. 2010, Brooks 2017). The third criterion assesses tangible linkages that physically connect biological and socioeconomic outcomes. Alignment of ecological and economic success is most likely when conservation activities yield direct socioeconomic benefits to the local community by enhancing ecosystem services that improve nutrition, health, security and well-being or present income-generating opportunities (Walpole and Wilder 2008, Brooks et al. 2013). Our fourth criterion considers local perception of the interdependence between conservation and socioeconomic goals, because win-win scenarios are more likely where people recognize or believe that socioeconomic progress requires sustainable natural resource management. Previous authors have addressed this by noting the importance of ecological awareness (Waylen et al. 2010, Gurney et al. 2014, Brooks 2017), but local understanding of ties between socioeconomic and biological goals may be equally important where linkages are artificial rather than physical. Finally, our fifth criterion examines investment of project-derived socioeconomic gains in conservation. Gains may be monetary or relate to increases in time, knowledge, skills and institutional capacity generated by the community conservation initiative that is invested in strengthening the initiative's own conservation aspects or promoting ecological awareness and community conservation elsewhere. Such knowledge sharing has been recognized as an important indicator regarding the scalability of projects considered for the Equator Prize (A. Fishman, G. Haber, F. Hervas, E. Winata, unpublished manuscript).

Case study on the Wechiau Community Hippo Sanctuary

The Wechiau Community Hippo Sanctuary (WCHS) lies in north-eastern Ghana along 36 km of the Black Volta River, bordering Burkina Faso to the West (Fig. 2). Founded in 1998 by the Paramount Chief of Wechiau and his sub-chiefs in lieu of a proposed government-run hippopotamus reserve, this 212 km² community protected area (CPA) comprises 17 communities.

Figure 2

The Wechiau Community Hippo Sanctuary and surroundings (left), connectivity map displaying locations of all protected areas along the Black Volta River between Bui and Boura dams (middle), and 2016 river survey of hippopotamus occurrences (right).

The sanctuary encompasses two zones: a 1–2 km wide core zone along the Black Volta River free of human habitation with bylaws to ensure minimal human impacts on hippos and their habitat, and an adjacent 5–10 km wide development zone that includes human settlements and farmland. Fourteen households originally located in the core zone were resettled to the development zone in 2002. Four ethnic groups differing in language, religion, and traditional lifestyle inhabit the development zone: Wala, Birifor, Hausa, and Dagaabe. The Wala, as original settlers of the land, hold all traditional authority and land rights (Sheppard et al. 2010).

Early phase socioeconomic developments in WCHS included targeted infrastructure improvements (wells, sanitation, solar lights, additional schools) and an ecotourism venture centered on hippo safaris (Sheppard et al. 2010). An organic shea nut harvesting cooperative was formed in 2008 as a second income source, with shea nuts hand processed locally into butter since 2015 in a new community-owned factory.

For our longitudinal case study of WCHS, we evaluated the initiative's performance 10 yr after its inception by applying SPECCS to data spanning 1999–2009 as published in Sheppard et al. (2010) on trends in hippopotamus numbers, bird diversity, infrastructure development, scholarships, the sanctuary's legal status, employment, finances, environmental awareness, and attitudes among school children and disadvantaged residents. We also collected comparable data 2010–2017 to assess performance 20 yr after inception, expanding some analyses to include comparative data from outside the WCHS and adding novel data sets on environmental awareness and attitudes among adults, staff and members of the organic shea cooperative. Data from interviews and questionnaire were analysed using QSR International's NVivo 10 Software (QSR International 2015).

Results

For each criterion, we report a 10-yr and 20-yr performance score and in brackets the corresponding quality scores before providing background on the data and analysis that aided evaluation. Further details on the data and rationale used in scoring are available on the Figshare data repository (see Data Availability) and in Supporting Information (Data S1), which can be uploaded in the online tool as an example evaluation (available online).2

Biological benefits

Trend in target of protection

Scores: 10-yr = 2 (4); 20-yr = 3 (4). We used hippo census data to evaluate this criterion. Earthwatch Institute volunteers (2000–2004) and trained local staff (2004–2017) conducted quarterly counts of the hippopotamus population within WCHS. Counts were conducted from canoes between 07:00 and 10:00, with canoes launched concurrently from four locations to cover four adjacent transects that jointly span the 36 km of river inside the sanctuary. Two staff per canoe verified each other's observations. We examined trends over time in the maximum total count per year using a generalized linear model (GLM) with a Poisson error structure. Sheppard et al. (2010) concluded that the local hippopotamus population was stable, averaging 14 hippopotami across years. In 2016, our Poisson model indicated an increasing trend from 2000–2017 (Appendix S1: Fig. S1), with an augmented annual average of 20 hippopotami.

Trend in threats to target of protection

Scores: 10-yr = 2 (2); 20-yr = 2 (2). Sheppard et al. (2010) found that the major threats to the local hippopotamus population were dramatically reduced given effective protection inside the WCHS from hunting, and habitat loss, but acknowledged risks associated with isolation and small population size. In 2016, the hunting ban and protection of core habitat remain effective within WCHS. Moreover, a 3-d survey along the Black Volta River in May 2016 from approximately 40 km north of the WCHS to its southern boundary (hereafter the 2016 river excursion) suggested that the hippo population may by now be larger and less isolated. Heading downstream in a wooden canoe, observations were made of hippopotamus and fishing nets in the water, hippo tracks and feces along the river bank, and human activities, farming practices, and shoreline habitat (Fig. 2). Opportunistic informal discussions with residents along the river provided insights on residents’ beliefs, fishing and farming practices, and recent hippopotamus observations and killings. Three hippos and 18 hippo exits (15 on the Ghanaian, 3 on the Burkina Faso side) were observed along the river bank between the Zukpiri Paramountcy and the northern border of the hippo sanctuary, and residents reported 9 additional recent sightings (Fig. 2). Observations of gold-panning activities upstream of the sanctuary suggested that water quality might be at risk. Moreover, lack of regulations outside the WCHS left the local hippo population more exposed to hunting in 2016 than 2009, with two hippos confirmed killed in 2011 and possibly one more in 2015. Given solid protection within the sanctuary, however, hippos nonetheless seem less vulnerable overall in 2016 than before the sanctuary's inception.

Habitat quality for taxon targeted for protection

Scores: 10-yr = 3 (2); 20-yr = 2 (3). We had two sources of data to evaluate this criterion. Anecdotal observations of river-bank habitat during the 2016 river excursion (see criterion 2), and habitat classification data collected in 2009 and 2016 along seven 2-km transects (four inside and three outside the sanctuary) used for bird surveys (see criterion 4). Each transect encompassed eight 250-m segments whose habitat type was classified as flood land, riparian forest, savannah, farmland, human infrastructure, and mixtures thereof. We calculated habitat degradation as the percentage of segments that included farmland or human infrastructure, then used two-sided unpaired t tests to compare degradation and change therein 2009–2016 inside vs. outside the WCHS. In 2009, bird transect habitat data indicated a mean percent degradation of 12.5% inside and 33% outside the WCHS (t = −1.581; P = 0.196) with habitat quality likely improving inside the WCHS given removal of human settlements from the core zone. The 2016 river excursion indicated that the WCHS’ core zone was pristine relative to other stretches of river, which often bore evidence of bank farming and logging. Habitat data from bird transects confirmed these observations, but also suggested that habitat degradation had increased between 2009 and 2016 both inside and outside the WCHS, with increases of 6% vs. 25%, respectively (t = −0.512; P = 0.657) for a mean degradation level of 18.8% on inside and 58.3% on outside transects (t = −1.224; P = 0.319).

Umbrella benefits to biodiversity or natural capital more widely

Scores: 10-yr = 2 (9); 20-yr = 2 (9). We used data on bird species richness to evaluate this criterion. A highly-trained community member, one of Ghana's preeminent bird guides, conducted bird surveys along four permanent transects inside and three permanent transects outside the sanctuary during both the wet and dry season each year from 2007 to 2017 (except 2009). Each transect was 2 km long with eight 250-m segments. During survey sessions, the bird guide walked all seven transects within 3–4 weeks, starting each between 05:00 and 05:45 and spending 30 min per segment to record all birds seen (with binoculars) or heard. Transect and segment-specific detection rates were calculated by conducting repeat surveys on each transect on consecutive days and dividing the number of species detected on both days by the total number of species recorded along the transect or segment in the two-day period. We generated species accumulation curves in R using the package vegan (Oksanen et al. 2017), adding sites in random order over 100 permutations. We used GLMs with a Poisson error structure to investigate the relationship between species richness and season (wet vs. dry), place (inside vs. outside the sanctuary), year, and all pairwise interactions, applying an information theoretic framework and model averaging (Burnham and Anderson 2002) to test all possible combinations of the parameters. We used generalized linear mixed-effects models (GLMMs) with transect nested in year as random effect and a Poisson or binomial error structure, respectively, to examine the relationship between species richness or detectability and habitat type.

A total of 167 bird species were recorded 2007–2009 (156 inside the WCHS; 120 outside), with greater bird species richness inside the sanctuary than outside the sanctuary during the dry season (Sheppard et al. 2010). By 2016, the total number of bird species observed rose to 227 (209 inside; 178 outside). Average species richness per transect 2007–2016 was significantly higher inside the sanctuary in both seasons (inside-wet 52.21 ± 6.27 [mean ± SD], outside-wet 44.42 ± 4.95, Wilcoxon's W = 2813.5, P < 0.001; inside-dry 56.06 ± 5.28, outside-dry 45.62 ± 5.44, W = 3289, P < 0.001), although species accumulation curves suggested that differences were strongest in the dry season. These results held when data were restricted to any three rather than all four inside transects, confirming that higher richness inside the sanctuary did not reflect greater survey effort. In addition, results were not biased by detectability, as year-round and season-specific detection rates were higher outside the sanctuary (mean percentage of species seen on both days was 60 ± 5 inside and 64 ± 6 outside, W = 2,205, P < 0.05).

GLMs supported these results, with the three top models (the change in the Akaike information criterion due to that model, delta AIC_c < 2) consistently including place, season, year, and the interaction between place and year, with the addition of interactions between season and place and season and year in the second- and third-best model, respectively. Only the model-averaged parameter for place, however, excluded zero from its 95% confidence intervals. Nonetheless, the interactions suggested a larger difference in species richness between inside and outside the sanctuary in the dry season and over time (Appendix S1: Fig. S2). The GLMMs found species richness to be lower but detectability to be higher in habitats commonly encountered outside the WCHS (e.g., human inhabited, savannah), suggesting that habitat type but not detectability may drive the observed trends. More species generally occurred in natural than human-altered habitat, with 22 species seen on average in transect segments with riparian forest vs. 19 in floodplain segments, 17 on savannah segments, 16 on farmland segments, and 15 near human infrastructure.

Biological resilience

Connectivity/viability of the target of protection

Scores: 10-yr = 1 (1); 20-yr = 1 (1). To evaluate connectivity in 2009, we searched the worldwide web, scientific, and gray literature for hippopotamus sightings and details on two protected forests in the neighboring country of Burkina Faso, with additional clarifying information provided by staff at that country's National Parks Department: Koulbi classified forest, located directly across the river from WCHS, and Bontioli classified forest, north of the WCHS on a tributary of the Black Volta River (Fig. 2). For our 20-yr evaluation, we undertook the 2016 river excursion (see criterion 2), obtained reports from a 2010 meeting between officials from Ghana (Wechiau and Zukpiri) and Burkina Faso discussing the potential for shared land and riverscapes to help hippopotamus and other wildlife, plus examined recent Ministry of Environment reports and clarifying information from the Directeur Général CEEF-BGA in Burkina Faso.

In 2009, connections between the WCHS's hippo population and other potentially persisting populations further downstream in Ghana and upstream in Burkina Faso were uncertain but possible (Sheppard et al. 2010). The 2016 river excursion and baseline hippo surveys in 2017 in nascent community resource management areas upstream of the WCHS in the Dorimon and Zukpiri Paramountcies (D. Millar, unpublished report) revealed that hippos now range to at least 40 km north of the WCHS. This increases the likelihood of population viability and connections between hippos in the WCHS and any that may remain in Bontioli or elsewhere upstream. Moreover, formation of the Dormion and Zukpiri community protected areas promises a contiguous area of protected riverine habitat along the Ghanaian banks of the Black Volta River extending to approximately 50 km north of the WCHS's northern boundary. In contrast, reinvigorated habitat protection on the Burkina Faso side of the Black Volta river, envisaged in 2009 for Koulbi and Bontioli Forests and along a wildlife corridor in between, has not so far materialized (André Jean de Dieu Ilboudo, personal communication).

Structural diversity of the species targeted for protection

Scores: 10-yr = 2 (2); 20-yr = 2 (2). We focused on the age structure of the hippo population to evaluate this criterion. During quarterly hippo censuses (see criterion 1), survey staff distinguished adults from juveniles by size. Because this method can be imprecise, we had reliable information only on the occurrence but not the relative abundance of age classes. Juveniles occurred in all years 2004–2017.

Project control over threats

Scores: 10-yr = 2 (1); 20-yr = 1 (1). In 2009, the hippo population was believed to predominantly reside within the WCHS. Likely too small to be viable in isolation over the long term, the population was considered demographically vulnerable, but effectively protected from hunting and habitat loss (Sheppard et al. 2010). With evidence in 2016 of hippos occurring beyond the boundaries of the WCHS (see criteria 2 and 5), complete protection of the local hippo population from hunting and habitat loss is no longer guaranteed. Although some communities north of the sanctuary have similar taboos against hunting and eating hippos as the Wala in the WCHS, others, notably those directly north of the sanctuary, are hostile to hippos given the threat hippos pose to fishing nets, crops, and lives, leading to two hippos being killed in the area in 2011, and possibly one more in 2015. River-bank farms observed during the river excursion indicate both habitat loss and heightened likelihood of human–wildlife conflict outside the sanctuary's boundaries. Moreover, observations of gold-panning operations north of the WCHS raised concerns over potential upstream pollution that might affect the health of hippos even within the sanctuary. Encouragingly, ongoing efforts to establish and strengthen community protected areas in the Dorimon and Zukpiri Paramountcies, respectively, may ultimately lead to improved protection of both wildlife and habitat upstream.

Socioeconomic benefits

Access to amenities

Scores: 10-yr = 3 (6); 20-yr = 3 (6). To evaluate this criterion, we obtained data on infrastructure developments between 1998 and 2016 by questioning community members, leaders, and elders in May 2016 in all 17 settlements inside the WCHS and 100 settlements elsewhere in the Wechiau Paramountcy. Interviews were conducted in either Wale or Birifor and recorded the number and approximate completion date of new banks, post offices, police stations, government offices, medical facilities, schools, school meal programs, places of worship, restaurants, bars, tourist accommodations, fuel stations, roads, telecommunication towers, internet access points, electricity access, solar power, hand-dug wells, drilled wells, dams, piped water, wastewater canals, public toilets, fishery facilities, shea butter factories, and commercial shea nut markets. We then compared the rate of infrastructure development inside and outside the sanctuary for small (<1,000 people) and large (>1,000 people) settlements, with population estimates obtained from the Wa-West District Health Administration. Infrastructure data used in Sheppard et al. (2010) considered a smaller list of infrastructure developments obtained late in 2007 from the Paramountcy's four administrative area councils.

Widespread improvement in access to a variety of amenities occurred within the WCHS from 1999 to 2009, with evidence of greater infrastructure development rates inside than outside the sanctuary. Development surveys in 2016 suggested that these patterns were holding, with continued, faster addition of new infrastructure inside than outside the WCHS for both small and large communities (Appendix S1: Fig. S3).

Employment creation

Scores: 10-yr = 1 (4); 20-yr = 2 (4). To evaluate this criterion, we obtained estimates of full time, commissioned, and occasional staff numbers and their salaries from WCHS officials, and annual figures on the number and income of shea cooperative members and factory workers from Savannah Fruits Company, the wholesaler that coordinates organic shea butter production on the sanctuary's behalf.

By 2009, ~50 regular and 100 occasional jobs for a total population of 10,268 had been created in the WCHS since inception. By 2017, the WCHS provided salaried or commissioned employment for 2,030 (22%) of an estimated 9,415 residents (Appendix S1: Table S1). Shea cooperative surveys (see criterion 22) suggested that the income women received from nut collection or factory work was a substantial contribution to total household income and welfare.

Capacity building

Scores: 10-yr = 3 (1); 20-yr = 3 (1). To evaluate capacity gains, we relied on anecdotal observations and data on scholarship recipients provided by the WCHS management. Scholarships to help brilliant students in financial need attain a high school education and beyond have been supported annually since 2005 by the Friends of Wechiau Group (matching funds from the WCHS education fund doubled the number of supported students 2005–2011). All scholarship students must serve the WCHS upon completing high school. By 2009 and 2017, 20 and 42 scholarship students, respectively, had attained secondary, and in four cases, tertiary education, subsequently assisting the WCHS management as capable tour guides, information officers and research aides. Moreover, exposure of WCHS residents to participatory decision-making and village representation on the WCHS management board in both evaluation periods has yielded community-wide benefits. Originally implemented to facilitate discussion of sanctuary matters, the decision-making structure proved critical in facilitating the implementation of the WCHS's commercial-scale organic shea venture, which in turn has led to widespread capacity building among women, specifically, given associated training, environmental education, and income.

Fostering cultural diversity

Scores: 10-yr = 3 (1); 20-yr = 3 (1). The WCHS initiative has since inception been built on traditional culture and beliefs. Local Wala and Birifor each have legends surrounding hippos that encourage their protection (Sheppard et al. 2010). The sanctuary's eco-tourism venture authentically showcases local cultural diversity with town tours offering insights into Wala culture, Lobi compound tours, and dance troupes highlighting Birifor culture, and river safaris building on the artisanal skills of Haussa residents. The organic shea cooperative respects traditional cultural activities and gender roles.

Socioeconomic resilience

Community rights to land and resources

Scores: 10-yr = 2 (4); 20-yr = 3 (4). Evaluation of this criterion was helped by documentation issued by local and national government bodies. For the 10-yr evaluation, we considered the WCHS to have secure de jure and de facto land but only de facto natural resource management rights. Traditional authorities such as the Paramount chief of Wechiau and his sub-chiefs hold land in trust for their people, with traditional land tenure recognized under Ghanaian law. Moreover, community influence over resource management was apparent given representation of each village on the WCHS management board, and evident respect for management board decisions by local government authorities. By the 20-yr evaluation, both land and resource management rights were established de jure and de facto because the WCHS is legally recognized since July 2010 as a Community Resource Management Area (CREMA). Consequently, authority over the management and utilization of natural resources now rests with the WCHS management board rather than Ghana's Ministry of Lands and Natural Resources (Asare et al. 2013).

Capacity to enforce project rules and regulations

Scores: 10-yr = 2 (1); 20-yr = 2 (1). Both official government documentation (see criterion 12) and a survey among WCHS staff helped evaluate this criterion. Seven long-term staff or management board members of the WCHS completed questionnaires in May 2016. The questionnaires asked respondents to identify opportunities and challenges posed by the sanctuary at the community, family, and individual level and to describe the approach to and success with addressing challenges. Furthermore, respondents were asked to express their opinions about the sanctuary and its management and their hopes and fears for the future.

For both evaluation periods, the WCHS's bylaws enjoyed legal backing, and rangers in several instances had handed violators over to police with prosecution pursued in court. Rangers are employed by the WCHS management board, such that enforcement is completely under local control. Limited fiscal and logistic resources, however, mean that patrols are often infrequent and geographically patchy. Moreover, lingering disputes over rosewood harvests and charcoal production suggest that enforcement and tightening of restrictions are more challenging when traditional authorities have vested interests in contravening them.

Decision balance between the local community and external stakeholders

Scores: 10-yr = 1 (1); 20-yr = 1 (1). To evaluate this criterion, we relied on both the staff questionnaires (criterion 13) and authors’ intimate knowledge about sanctuary dealings. At the time of our 10-yr evaluation, the WCHS had transitioned from at first relying heavily on an on-site external sanctuary advisor (author D. J. Sheppard) for day-to-day management to operating more independently but with continued, regular guidance from the external advisor, who in 2006 had relocated to Accra, a day's journey away. In 2010, the advisory role was successfully passed to a local, but externally funded, sanctuary advisor, increasing autonomy. In 2015, misunderstandings between two external partners led to the advisor being dismissed without replacement and without due consultation of the WCHS management board. Lack of an advisor has, on the one hand, forcibly increased decision-making by the WCHS management board, but on the other hand has left them feeling disadvantaged in stakeholder negotiations around, for example, accessing funds held in trust for the WCHS by external project partners, and setting tenancy terms for the sanctuary's shea factory.

Equitability of benefit sharing

Scores: 10-yr = 3 (1); 20-yr = 3 (2). Infrastructure development (see criterion 8), employment (see criterion 9), and capacity building (criterion 10) in both evaluation periods had a wide geographic reach and included all ethnic groups. Infrastructure developments such as clean water access, schools, and clinics also ensured that the age and gender composition of beneficiaries was diverse. Gender-diversity of employment has since been reinforced by the organic shea venture, which given local cultural norms, primarily benefits women. Moreover, key positions in the sanctuary's management board, originally filled by traditional Wala chiefs, have since been passed on to ordinary community members more diverse in ethnicity and gender.

Economic sustainability

Score: 10-yr = 3 (4); 20-yr = 3 (4). Three sources of information helped evaluate this criterion: (1) financial records from the WCHS management board detailing the sanctuary's tourism revenues; (2) data on external monetary and in-kind donations supplied by the Calgary Zoological Society, Canada, and the Nature Conservation Research Centre, Ghana; (3) information from Savannah Fruits Company on the factory rent and conservation premiums paid to the WCHS's conservation trust fund, which is administered externally but intended to help the community maintain and enhance conservation measures and associated, ecologically sustainable business ventures.

Based on continued external support from a diverse set of external donors, and sufficient internally generated revenue to cover the salaries of sanctuary staff (see employment data described in criterion 9), we had considered financial viability excellent in 2009 (Sheppard et al. 2010). In the 20-yr evaluation, project-generated income to the WCHS consistently matched or exceeded staffing costs since 2004. Data on visitor number and tourism-related income, however, reveal the risk associated with single income streams: tourism-related income declined dramatically 2014 onward, likely due to the Ebola crisis elsewhere in West Africa discouraging travel (Appendix S1: Table S2). By then, however, the WCHS had a second income stream from its organic shea venture, and so was buffered financially, at least theoretically. Lack of established procedures to access shea-generated funds held in trust by an external partner nonetheless meant that staff were left unpaid for several months, a fact reflected in low decision-balance scores above.

Income diversification

Score: 10-yr = 3 (4); 20-yr = 3 (2). Evaluation of this criterion was informed by employment data (see criterion 9) and financial records (see criterion 16 and Appendix S1: Table S2). Since its inception, the WCHS initiative has had long-term support from a diverse set of donors and in both evaluation periods enjoyed direct income from eco-tourism and shea.

Social capital to problem solve

Scores: 10-yr = 3 (1); 20-yr = 2 (1). To evaluate this criterion, we relied on both the staff questionnaires (criterion 13) and authors’ intimate knowledge about sanctuary dealings. At its 10-yr evaluation, improved problem-solving capacity was illustrated by the sanctuary's representational management structure, community leaders’ foresight to pursue shea as a second income stream, and the chiefs’ demonstrated commitment to be watchful over new settlers and new developments once concerns over a possible population influx were raised by external partners (Sheppard et al. 2010). By 2016, the WCHS management and its constituent communities had demonstrated additional foresight in implementing preemptive burns to prevent bushfires that might harm the core zone. Moreover, with rising charcoal production in the wider region, the management board attempted to firmly discourage local participation in this environmentally detrimental practice by requesting traditional leaders to ban charcoal transports through the sanctuary, albeit unsuccessfully to date. Responses to other challenges, however, occurred only once initial damage was done. When communities north of the sanctuary killed hippos in 2011, sanctuary staff ensured that the carcasses were buried rather than eaten to discourage further poaching. Similarly, when it was discovered in 2015 and 2016 that rosewood was being harvested in the core zone in clear violation of the bylaws but possibly with permission from one of the traditional chiefs, the actors responsible were taken to court and the timber was left to rot. Livestock transitioning through the core zone to access the river for water has been a long-standing problem, and yet to date only one small damn for alternative water access has been put in place. In both 2015 and 2016, organic shea nuts from several sanctuary communities were contaminated by insecticides sprayed to control malaria, indicating that the community possessed insufficient capacity to navigate the complexities of an adequate compromise. Poor shea harvests in 2016 also raised concern over the long-term sustainability of commercial shea exploitation in the area, a factor not previously considered.

Linkage mechanisms

Ecological awareness

Scores: 10-yr = 1 (6); 20-yr = 2 (9). To assess local environmental awareness, student surveys were undertaken with a representative sample of primary and middle school students from inside WCHS in 2004 (200 students, 7 schools), 2007 (159 students, 6 schools), 2011 (290 students, 14 schools) and 2016 (194 students, 20 schools). In 2016, we additionally surveyed 199 students from 18 schools outside WCHS. Students were selected randomly but stratified based on school size and gender ratios. Questions tested their ability to describe plants and vertebrate animal classes (basic ecology), photosynthesis and animal behaviors (ecological linkages), nature's benefits to humans, conservation concepts, and the WCHS's purpose and benefits. Each student was interviewed in their preferred language. Answers were recorded verbatim, then categorized to assign a percentage score for individual questions, sections, and the entire questionnaire. Results from inside the sanctuary were analysed for temporal trends while also accounting for student age, gender, grade, and tourism exposure (binary). For 2016 surveys, we additionally compared results from schools inside and outside the WCHS while accounting for age and gender. We used GLMs with a binomial error structure to analyze all possible combinations of the covariates as additive effects and calculated model-averaged parameter estimates in R (R Core Team 2016) using the MuMIn package (Barton 2016).

Further insights on environmental awareness and attitudes came from standardized interviews with adults conducted in May and June 2016 in all 17 communities inside the WCHS and 17 randomly chosen communities outside. Respondents included each community's leader where possible, but were otherwise picked randomly in proportion to the community's population size, with 153 residents inside and 200 outside interviewed in their language of choice—English, Birifor or Wale. Questions concerned uses and stewardship of the environment, environmental change, and understanding of and opinions about conservation.

For the 10-yr evaluation, student surveys suggested that environmental knowledge among school-age children in the sanctuary was cursory. They had some understanding of their environment, albeit biased toward agricultural rather than undisturbed settings, with average scores for basic ecology and ecological linkages at 78% and 41%, respectively. Their understanding of conservation was mixed, with 91% aware of the sanctuary, 59% able to describe its key features, but only 15% capable of articulating the meaning of conservation or extinction.

Expanded student surveys in the 20-yr evaluation mimicked earlier results suggesting that boys and older students generally scored higher, except when describing conservation concepts (Appendix S1: Fig. S4a). While knowledge about basic ecology, ecological linkages, and key features of the WCHS declined over time, students’ understanding of nature's benefits to humans and conservation concepts improved (Appendix S1: Fig. S4a). Students inside the sanctuary generally scored higher in all categories except, ironically, their understanding of conservation concepts, which was approximately equal inside and outside the sanctuary (Appendix S1: Fig. S4b).

Among adults, understanding of conservation concepts was good in 2016 with many (41% inside, 41% outside) describing conservation as “protecting the environment” or “protecting wild animals and plants,” and all but one participant stating that humans had a role to play in protecting nature, specifically by planting trees, or stopping bush-burning, tree felling, and pollution. Most (91% inside, 87% outside) had observed environmental changes such as loss of soil fertility, deforestation, and altered rainfall and temperature patterns. Many (35% inside, 56% outside) also predicted future deterioration, with several (71% inside, 56% outside) suggesting that action was needed to prevent this. Common examples of benefits gained from protecting the environment included necessities such as food, clean water, and good health or disease prevention. Participants explained extinction as a species or all its members being dead or far away (85% inside, 51.5% outside), but provided more incorrect examples of extinction (93% inside, 95% outside) than correct ones (87% inside, 72% outside). They attributed species loss to hunting, bush burning, and deforestation. Knowledge of key features of the WCHS, such as its purpose of protecting the environment, hippos, or other plants and animals, was higher inside (86%) than outside (59%) the sanctuary, with 30 participants living outside the WCHS never having heard of it.

Attitudes and emotional investment

Scores: 10-yr = 3 (2); 20-yr = 3 (4). Student surveys and interviews with adults (see criterion 19) helped evaluate this criterion in addition to interviews with inconvenienced sanctuary residents. These are residents of the nine communities bordering the sanctuary's core zone, who prior to the sanctuary's establishment lived inside the core zone or depended on its natural resources. In 2007, 46 participants were interviewed and asked about their knowledge of the sanctuary, associated benefits, what they believed would happen if the project failed, and any additional opinions or concerns. In 2016, the same participants were interviewed again and additionally asked what sacrifices they made for the sanctuary. Where original respondents had passed away (six cases), relatives of the deceased were interviewed.

For the 10-yr evaluation, Sheppard et al. (2010) reported a palpable sense of pride among community members and the finding that even among directly inconvenienced residents, only 33% raised any grievances, most (72%) acknowledged benefits, and most (83%) thought that discontinuation of the sanctuary would hurt the community.

Repeat surveys with inconvenienced households in 2016 yielded similar results, with 72% acknowledging benefits and 89% foreseeing negative impacts should the sanctuary discontinue. In describing the sanctuary, however, respondents more frequently focused on restrictions in 2016 (52%) than in 2007 (30%) such as no farming, tree felling, hunting, bush burning, and grazing. Directly asked, almost all (96%) 2016 respondents mentioned sacrifices made for the sanctuary, including above restrictions, prohibitions on oyster harvests and displacement. Nonetheless, only 21% raised grievances.

Among the wider community, most adults participating in the standardized interviews (see criterion 19) recognized that the WCHS brought benefits (100% inside, 92% outside), such as infrastructure and employment, while at the same time being aware of sacrifices (93% inside, 56% outside) such as restrictions on farming and hunting. The vast majority supported continuation of the WCHS (99% inside, 79% outside).

Similarly, among those interviewed during shea cooperative surveys (see criterion 22), almost all (99%) thought the shea cooperative was positive for the community, with 88% specifically expressing gratitude for the shea factory, and many naming benefits of increased income (72%) and consequent improvement in access to education for their children (57%), healthcare (36%), and life overall (19%).

Finally, staff questionnaires (see criterion 13) indicated that the staff's own perception and their impression of the community's perception toward the project were consistently positive, although several raised concerns over community members disregarding bylaws (85%) and the need to ensure continued, equitably distributed socioeconomic benefits (28%).

Tangible linkage between biological and socioeconomic outcomes

Scores: 10-yr = 3 (1); 20-yr = 3 (1). For both evaluation periods, the sanctuary's two primary income streams were directly tied to environmental protection, with tourist revenue dependent on successful protection of hippos and their habitat, and shea collectors paid extra for adhering to organic practices (see employment data and financial records described in criterion 9 and 16, respectively).

Local perception of the interdependence between biological and socioeconomic outcomes

Scores: 10-yr = Not collected (0); 20-yr = 3 (2). We had no data on the local perception of linkage for our 10-yr evaluation. For the 20-yr evaluation, insights were gleaned from interviews with members of the shea cooperative. We interviewed 235 women (141 Birifor and 94 Wala women) across 11 settlements within WCHS involved in the organic shea butter venture. Of these women, 59 worked in the factory, the remainder harvested shea nuts. Interviews were conducted in the women's language of choice (Birifor or Wale) by two trained local women in May and June 2016, addressing questions about harvesting practices before and since establishment of the WCHS shea cooperative, shea-derived income and its uses, and the women's understanding of linkages between the cooperative and nature conservation. Most respondents (77%) understood that the organic shea venture contributed to the protection of the local hippos given its “forest friendly” organic nature and absence of impingement on hippos or their habitat. Even more (86%) mentioned that the hippos attract foreign development partners and so help facilitate ventures like the shea cooperative. They therefore recognized both a direct and indirect linkage, but perhaps not the direct dependence of income (e.g., organic shea premiums) on successful environmental protection.

Investment of project-derived socioeconomic gains in conservation

Scores: 10-yr = 3 (4); 20-yr = 3 (4). To assess investment in conservation, we considered both monetary investments and time spent by community leaders and sanctuary staff on promoting conservation elsewhere. Financial records (see criterion 16) allowed us to calculate the percentage of total revenue generated from the ecotourism (visitor income) and shea factory ventures (shea factory rental and conservation premiums) invested into the WCHS's conservation trust fund and salaries of the sanctuary's conservation staff (the manager and his rangers).

At the 10-yr mark, investment in the conservation trust fund averaged 2% annually, whereas conservation salaries amounted to 54%, for a total conservation investment of 54% annually on average. Moreover, the sanctuary's Chairman had travelled to two other regions in Ghana to advocate for community conservation. By the 20-yr evaluation investments into the conservation trust fund had improved to 15% and salary investments dropped to 25% for a total average annual conservation investment of 63%. Moreover, sanctuary staff completed two separate visits to Kenya to encourage community conservation initiatives there.

Overall progress

The WCHS maintained high total performance scores in both the 10-yr (76%) and 20-yr evaluation (76%), with the quality of evaluation data increasing from 30% to 34% (Table 2). While socioeconomic benefits and linkage mechanisms increased, the performance score for biological benefits and socioeconomic resilience remained unaltered and biological resilience declined from 56% to 44%. This reduction is the bittersweet outcome of success; threats to hippopotami have become harder to control because the animals have spread beyond sanctuary boundaries, likely because of population recovery given effective protection within the WCHS or possibly an influx of hippopotamus from Bui National Park given loss of suitable habitat there after completion of Bui damn.

Table 2. Summary of performance and evaluation quality scores for the Wechiau Community Hippo Sanctuary 10 and 20 yr after project inception

Themes	10-yr evaluation (%)		20-yr evaluation (%)
	Performance	Quality	Performance	Quality
Overall project score	76	30	76	34
Biological benefits	75	47	75	50
Biological resilience	56	15	44	15
Socioeconomic benefits	83	33	92	33
Socioeconomic resilience	81	25	81	27
Linkage mechanisms	83	29	87	44
Total biological score	65	31	60	32
Total socioeconomic score	82	29	86	30
Total benefits score	79	40	83	42
Total resilience score	68	20	63	21

Notes

• Scores are summarized as percentage scores overall, for the five elements of success (biological benefits, biological resilience, socioeconomic benefits, socioeconomic resilience, linkage mechanisms), and for four overarching themes: total biological score, total socioeconomic score, total benefits score, total resilience score.

The evaluations also highlight that the project, while successfully pursuing both its ecological and socioeconomic goals, is currently somewhat stronger on the socioeconomic side and in immediate benefits rather than longer-term resilience (Table 2, Fig. 3).

Figure 3

Diagram illustrating the balance of performance (solid lines) and of evaluation quality (dashed lines) among five elements of success for community-based conservation at the Wechiau Community Hippo Sanctuary 10 (blue) and 20 yr (orange) after its inception.

Discussion

To our knowledge, SPECCS comprises the most comprehensive list of locally applicable success criteria for community-based conservation published to date. Its novelty lies in offering a standardized format for assessing both performance and evaluation quality. By facilitating comparison within and among projects, SPECCS answers calls for more comprehensive and frequent success monitoring (Agrawal and Reford 2006, Brooks et al. 2013, Baynes et al. 2015, Oldekop et al. 2016, Brooks 2017).

Our tool incorporates many of the criteria identified as important components of success in previous literature (Walpole and Wilder 2008, Waylen et al. 2010, Gurney et al. 2014, Baynes et al. 2015, Brooks 2017, Reid et al. 2017). In addition to a comprehensive evaluation of both biological and socioeconomic outcomes, SPECCS uniquely incorporates the linkages between these two themes in line with a social-ecological systems framework (Cumming and Allen 2017) and also distinguishes shorter-term benefits from longer-term resilience mechanisms.

SPECCS thus aligns with the emphasis on resilience and adaptability among criteria for several international funding awards (Equator Initiative 2017, UK Government 2017; A. Fishman, G. Haber, F. Hervas, E. Winata, unpublished manuscript) and recognizes that poverty reduction is rarely achieved through economic improvements alone (Gurney et al. 2014). Poverty is multi-dimensional, with literacy, health, disempowerment, and vulnerability all playing a part in addition to income (Agrawal and Reford 2006). Comprehensively, addressing poverty therefore involves improvements in (1) economic opportunity, as measured in our tool by employment creation, access to amenities, and capacity building; (2) empowerment, reflected in the devolution of land or resource management rights, enforcement capacity, equitable benefit distribution, and strengthened local decision-making powers; and (3) security, represented in our evaluation tool by income diversification and social capacity to problem solve (The World Bank 2001, Gurney et al. 2014). Aspects of poverty alleviation have sometimes been considered as predictors of other outcomes (Brooks et al. 2013, Baynes et al. 2015, Oldekop et al. 2016), whereas we treat them as outcomes in their own right.

By evaluating success on a four-point scale, we are able to acknowledge even limited progress. For example, we give full marks if community conservation initiatives create employment for 25% of the community's population and good marks (two out of three) if the target of conservation shows stable trends, whereas Brooks and colleagues coded equivalent scenarios as limited success and outright failure, respectively (Brooks et al. 2013, Brooks 2017). Our apparent leniency abandons success as a binary concept, instead focusing on incremental progress in full recognition that success for community-based conservation initiatives ultimately is multi-faceted and dynamic (Agrawal and Reford 2006, Brooks et al. 2013, Baynes et al. 2015, Oldekop et al. 2016, Brooks 2017).

SPECCS acknowledges progress not only in performance but also evaluation quality by systematically accounting for variability in the quality and completeness of appraisal data. Data quality and its variability have been considered before in the context of meta-analyses looking at factors that might influence success (Brooks et al. 2013, Oldekop et al. 2016, Brooks 2017), but have not to our knowledge been developed for adoption on a project-specific basis. By incorporating a criterion-specific quality measure, the need and potential investment opportunity for community-based conservation monitoring data are highlighted.

Based on our example evaluation, the WCHS (winner of the 2008 United Nations Equator Prize) is sustaining success two decades after inception. This is an impressive feat given that many initiatives that appear successful to start crumble once external start-up funding ceases (Gurney et al. 2014). WCHS instead remains a beacon of hope illustrating that community conservation can foster sustained, positive changes for both people and nature. And yet even this celebrated project has considerable room for improvement. SPECCS pinpoints areas of strengths and weakness in both outcomes and monitoring quality. Biological outcomes have fallen behind socioeconomic outcomes, with weak and worsening biological resilience given limited connectivity and control over threats. Encouragingly, the development of additional community-based conservation initiatives underway north of the WCHS may alleviate both challenges. Among socioeconomic outcomes, a decision-balance favoring external stakeholders proved the weakest link. Efforts are now underway to redress the imbalance via a multi-party memorandum of understanding that highlights community empowerment as a central, unifying goal. This is illustrative of the power a comprehensive, standardized evaluation tool holds in motivating adaptive management. SPECCS should also encourage users to think carefully about the data required for effective evaluation. Quality of evaluation data for the WCHS improved over time, but still relied heavily on anecdotal and descriptive data.

Undoubtedly, SPECCS has limitations. Our criteria, for example, expect a single performance value for potentially complex issues such as the trend in threats. To pick a responsible score in such situations, we encourage evaluators to separately undertake a basic risk analysis integrating the relative likelihood and severity of various factors. Where multiple indices or data sets can be used to address a question, it may also be difficult to determine evaluation quality, but we suggest as a simple rule of thumb that analytical and design quality scores should be based on the data and analysis that most conclusively supports the performance score assigned. Should evaluators instead choose to enter average data quality, the tool also accommodates noninteger quality scores. Finally, it is worth noting that the variable number of criteria per element means that total scores for some elements (e.g., biological resilience) are more sensitive to scores for individual criteria than others. Given above limitations, it is important that researchers and practitioners be as objective as possible when evaluating their initiatives, and carefully record the rationale used in scoring each criterion.

SPECCS is useful because it allows for rapid assessments of strengths and weaknesses and is applicable to a global conservation audience. The process is made easier through the development of an interactive spreadsheet and online web tool that automatically calculate and visually display the results of the evaluation. More than a simple indicator of success, the tool helps track progress to support adaptive management if repeatedly applied to individual initiatives. Much of the literature highlights the need for long-term commitment of support, especially by external stakeholders (Brooks 2017). Regular monitoring with our tool should help keep local stakeholders motivated and assist external stakeholders to effectively target their support. Award and grant-issuing organizations may also find the tool useful in prioritizing investments. Comparisons between initiatives might reveal trends and thereby help streamline how projects are set up, or highlight gaps that should be addressed. Repeated application across initiatives will ultimately yield data needed to better determine critical components of community-based conservation success on a global scale and thereby support attainment of the Aichi targets (Convention on Biological Diversity 2017) and UN Sustainability goals (Abensperg-Traun 2009).

Acknowledgments

We thank the communities in and around the Wechiau Community Hippo Sanctuary for their assistance, especially the Paramountcy Chiefs and CPA board members. We also thank all who helped gather or prepare data: Tungbani Issahaku Agba, R. Y. Abudulai Issahaku, Haruna Jebuni, Sammeh John, Dasana Joel, Saaka Hamida, Yaw-Bile Fatimata, Savannah Fruits Company, especially Diane Banuro and Felix Basing, Charles Bandari, and André Jean de Dieu Ilboudo. Laura Keating provided invaluable help with analyses and designed the web tool. Two anonymous reviewers offered helpful feedback. Funding was provided by the Calgary Zoological Society's Conservation Fund, Husky Energy and the Centre for Conservation Research.

Notes

1 https://calgaryzoolk.shinyapps.io/cbcmatrix_app/

2 https://calgaryzoolk.shinyapps.io/cbcmatrix_app/

Data Availability

Data are available from Figshare: https://figshare.com/s/d901595dc28c380ea20a