Mitigation of coral ecosystem service-related social issues: evidence from a coastal development project in a developing country

ABSTRACT

Biodiversity offset, with a focus on securing biological validity, is primarily practiced in developed countries. However, various social issues, such as the tradeoff between ecosystem conservation and the use of ecosystem services, have arisen and require attention. To address these issues, this study proposes an approach applicable at the planning stages of coral offsetting. We propose a comparison of the relationship between social characteristics and coral ecosystem services across offset candidate sites, suggest an analysis of restorability and the extent of human dependency on coral ecosystem services as part of the process of selecting an offset site, and confirm the practicability of our approach for a specific case of coral offsetting in a developing country. The results reveal that all candidate sites (even across a small area) have advantages and disadvantages regarding the mitigation of social issues. We further discuss the implications of these results with respect to the sustainable usage of endangered corals.

KEYWORDS:

• Biodiversity offset
• social issue
• ecosystem service
• coral
• developing country

1. Introduction

Biodiversity offsets (offsetting) are measurable conservation outcomes from actions taken to compensate for the residual adverse impacts of development projects on biodiversity after prevention and mitigation measures have been taken (BBOP (Business and Biodiversity Offsets Programme) 2012a). Offsetting is primarily practiced in developed countries and focuses on securing ecological validity (Gelcich et al. 2017). However, in developing countries, especially in rural areas, local livelihoods largely rely on natural resources and are lost or restricted with the designation of project areas. Therefore, international aid organizations have requested that executing agencies consider the social aspects in the planning of offsetting (BBOP (Business and Biodiversity Offsets Programme) 2012a; Ledec and Johnson 2016). Furthermore, the importance of incorporating socioeconomic factors into offset design is recommended by several scholars (Brownlie and Botha 2009; Gelcich et al. 2017). The benefits obtained from natural resources are generally referred to as ecosystem services (hereinafter referred as ‘ES’); these are classified into provisioning services, such as food; regulating services, such as flood control; cultural services, such as recreation; and supporting services such as nutrient cycling (Millennium Ecosystem Assessment 2003).

Impacts on ecosystems are usually treated as environmental considerations, in accordance with the mitigation hierarchy that defines avoidance, minimization, and restoration procedures (BBOP (Business and Biodiversity Offsets Programme) 2012a). However, the impact on ES has not been considered explicitly because of the lack of adequate guidance (Honrado et al. 2013). Impacts on ES may be indirectly considered as social issues, because the social items assessed in impact assessment often include the utilization of land and local resources, local economic factors (such as livelihood), and equality of benefits and losses (JICA (Japan International Cooperation Agency) 2010; The World Bank 2004); the status of these items often changes because of the ecosystem degradation accompanying development projects. However, the way in which ecosystems and ES are considered separately is inefficient and ineffective as it does not take into account interaction among them (Tallis et al. 2015). A fundamental interaction is the predominant supply of ES by healthy ecosystems; however, such ecosystems are degraded with the increasing use of ES (Moreno-Mateos et al. 2015). Therefore, it is necessary to develop a method to bring this tradeoff closer to a win–win relationship that encourages better social consideration of ES. However, the development of such a methodology should also consider more complex social phenomena other than the tradeoff. First, in the long term, offsetting could potentially restore ES loss through the recovery of ecosystem loss induced by development projects. Additionally, spillover benefits from the offset site on ES can be expected as the ecosystem recovers during the offsetting period. Yet there are also several ES-related issues that arise with the creation of an offset site. This includes uncertainty regarding the recovery of the ecosystems and corresponding ES, conflicts stemming from the intensive use of resources after achieving biodiversity gains, and additional ES losses because of restricted social activities for offset site management. Furthermore, development site ES losses remain during the offsetting period unless effective countermeasures are taken.

Recent material published by international aid organizations conceptually or philosophically describes the need for consideration of ES in development projects, but specific approaches that suit different ecological and social conditions are not described (Baker et al. 2013; Tallis et al. 2015). Regarding offsetting, there are very few cases in which ES and relevant social issues are considered and reported (Benabou 2014). Moreover, although some offset cases can be obtained from Environmental Impact Assessment (EIA) disclosures on donor websites, the ES consideration is not detailed (see EIA disclosure example of IFC (International Finance Corporation) 2017). Human interaction with ES is also variable, depending on the circumstances surrounding a particular development project (Wells and McShane 2004; Esteve et al. 2007). Therefore, understanding and analyzing the social characteristics of the people who utilize ES are important for offset planning.

Few case studies for marine ecosystem offsetting, especially in the topic of corals, have been reported. For example, a recent literature review of offsetting cases between 1999 and 2014 revealed that the majority of studies are related to freshwater environments (66%), followed by terrestrial environments (47%), whereas very few recent studies explored marine environments (1%) (Gonçalves et al. 2015). Moreover, only 17% of marine ecosystem offsetting cases are related to coral ecosystems (Jacob et al. 2018). In contrast, the development of coastal coral habitats is expected to be advanced (The World Bank 2016a); most coral reefs are notably located along the coasts of developing countries (Gemez, 1997). Despite this situation, coral restoration techniques remain underdeveloped (Hein et al. 2017), and critical information for evaluating restoration techniques is lacking (Hein et al. 2019). Additionally, many scientists recognize the uncertainty of the effects of specific restoration techniques, such as simple transplanting, transplanting of nursery raised corals, and electro-stimulation (Moberg and Rönnäback 2003; Precht et al. 2005; Edwards and Gomez 2007). Coral reefs provide humans who live near them with a variety of benefits and ES such as seafood, tourism, educational tools, and coastal protection (Gomez 1997; Hein et al. 2019); therefore, many people, particularly in developing countries, have been negatively impacted by the degradation of coral reefs.

In this paper, we propose an approach to selecting advantageous offset sites and subsequently identifying destructive activities that need to be regulated therein. This approach is expected to bring the relationship between coral ecosystems and ES closer to a win-win situation. Thus, the goal of the approach is not limited to achieving ‘no net ecological loss’ or ‘gain’ of corals; it also includes measures to mitigate social issues that vary depending on social characteristics and its relation to coral ES (hereinafter referred as “CES). Furthermore, we apply the approach to a wharf development project in a developing country where coral offsetting is required to confirm its practicality. In this project, coral transplantation was performed by agencies executing the wharf project, prior to construction, as a measure to mitigate the loss of 475 m² of corals due to reclamation. However, the survival rate of the transplanted coral decreased. As their prognosis is unknown, the project executing agency and the donor agency agreed to implement offsetting following a series of stakeholder consultations. The idea of the offset is to create and manage protected areas at a site where mechanically damaged corals exist in a community conservation area (hereinafter referred as ‘CCA’) under registration. By applying the above-mentioned approach and incorporating the traditional knowledge of local people in government initiatives, social issues relating to CES are expected to be effectively mitigated.

2. Methodology

2.1 Consideration approach

Griffiths et al. (2019) emphasized the necessity for offsetting to achieve social gains equivalent to incurred social losses. Social gains are theoretically expected as a primary outcome of offsetting. However, several issues related to ES may arise alongside offsetting and affect the achievement of social gains. Therefore, an approach that resolves these issues must be considered. As this study focuses on the planning stages of offsetting, issues that arise during site selection and the approaches considered for addressing each issue are shown in Figure 1. There are four corresponding issues: (1) uncertainty that an ecosystem and ES will recover, (2) tradeoffs between conservation and use of an ecosystem, (3) conflict caused by the intensive use of resources after achieving biodiversity gains, and (4) additional losses in ES use because of restrictions on relevant activities during the offsetting period. In the consideration approach, we address the first issue by exploring offset sites with higher restorability¹ and aim to effectively restore the ES from a biological and social perspective. To address the second, third, and fourth issues, we explore offset sites where people are less dependent on ES and identify destructive social activities to be regulated at an offset site to mitigate the negative impacts of establishing an offset site on use of CES.

Figure 1. Issues that may arise in an offset site and consideration approaches to address each issue

2.2 Research approach

The research required for the above-mentioned consideration approach and its prospective output are shown in Figure 2. The research approach in this study is developed to specifically target coral ecosystems. First, rapid assessment is proposed to select an offset candidate site. Afterward, biological surveys, such as underwater observations, and social surveys, such as structured interviews, are proposed to understand the biological state and social characteristics of each candidate site, respectively. To select an offset site with high potential restorability, the element of restorability needs to be analyzed with data on social characteristics (collected from the interviews) and bio-equivalency (calculated from the results of the biological survey). To select an offset site where people are less dependent on CES, the level of engagement² in social activities using CES (hereinafter referred as ‘CES activities’) needs to be analyzed using social characteristics. Because several types of ES (such as cultural services) cannot simply be replaced monetarily (Bullock et al. 2011; Calvet et al. 2015; Moreno-Mateos et al. 2015), an analysis of the level of engagement in CES activities is proposed to estimate human dependency, instead of calculations of monetary value. Additionally, biological state and social characteristics are used to analyze the correlation between CES activities and actual coral damage; the results are used to select the destructive CES activities to be regulated at the offset site. Notably, this study considers ‘human dependency on CES’ to be equal to the ‘level of engagement in CES activities.’

Figure 2. Research approach and prospective output

This research approach is eventually expected to produce two outcomes: the selection of a socially and biologically superior offset site and the formulation of an efficient offset site management plan, even though these outcomes are under the project’s scope. The specific procedure of the research approach is described below.

2.3 Procedure of the research approach

2.3.1 Candidate site selection

BBOP (Business and Biodiversity Offsets Programme) (2012b) proposes a set of criteria for better selection of offset candidate sites as follows: (1) exclusion of sites where bioequivalence with affected sites and ecological net gain cannot be expected and (2) selection of sites where social or political support is easily obtained. With reference to these criteria, we set the criteria for the selection of candidate sites (Table 1). First, to evaluate bio-equivalency and to achieve net gain, the existence of corals is a minimum requirement. Therefore, one of the criteria is that ‘there is a certain amount of corals.’ Second, to effectively achieve net gain, a site must be chosen that assures coral recovery. Because coral damage accompanied by chemical reactions, such as climate change and water pollution, is difficult to reverse with offset schemes, we focus on recovery from mechanical damage. Therefore, ‘there is a certain amount of mechanically damaged coral’ is a criterion. Third, to obtain the support for the offset site management, easy access to the offset site is a critical condition for users and managers. Therefore, ‘accessible for users and managers of offset site’ is a criterion. It is notable that, to obtain the social or political supports, coordination among the stakeholder is must but is not the scope of this study.

Table 1. Criteria set for the selection of offset candidate site and reasons for setting those criteria

Criteria	Reasons
There is a certain amount of corals	Existence of corals is minimum requirement.
There is a certain amount of mechanically damaged coral	To choose a site where achieving coral recovery is assured, it is better to focus on mechanically damaged corals.
Accessible for users and managers of offset site	For ease of obtaining the support for the management of offset site

2.3.2 Restorability analysis

Biological perspective

To secure the restorability of an offset site from a biological point of view, bio-equivalency between the affected site and the offset site needs to be measured. As biological measurement should be easily performed by non-professional offset executors, especially in developing countries (Gibbons and Freudenburger 2006), this study proposes locally available methods and parameters that can measure quality and quantity of coral ecosystems.

For benthic data collection, simple measurement equipment, such as a quadrat, is to be placed on the seafloor, which represents the areal coral condition. Subsequently, previously collected indicators for an impact assessment of the development project, such as coral cover, individual number of benthic invertebrate species, colony number of seaweed/algae, and depth are to be recorded. Regarding ambient environment, comparable indicators such as temperature, salinity, and the pH of surface water are to be recorded using a data logger. The Shannon–Weaver H’ diversity index can be calculated from the coral composition data by implementing the following formula, where ni is the ith species number, N is total number of species, and Pi is ni/N:

$H^{\prime }=-\sum PilogePi$

A cluster analysis using the PRIMER® software (version 6, manufactured by PRIMER-E Ltd) offers a potential method to understand the bio-equivalency between affected and candidate sites.

Social perspective

Kelleher (1999) raised the following criteria to ensure the practicality and feasibility of the Marine Protected Area

(MPA) in the IUCN guideline for MPA: (1) compatibility with existing uses, particularly by locals; (2) social and political acceptability, i.e. degree of community support; (3) degree of insulation from external destructive influences; (4) ease of management or compatibility with existing management regimes; and (5) accessibility for education, tourism, and recreation. By targeting the recovery of ecosystems and ES through the establishment of a protected area, our approach refers to these criteria to suggest the contents of a social survey for restorability analysis (Table 2). The contents of the social survey consist of items to be grasped and the corresponding questions. The level of each item determines the level of restorability. The item can be estimated with answers to the questions by the method described below in detail.

Table 2. IUCN’s criteria for ensuring the practicality and feasibility of the Marine Protected Area (MPA) and corresponding contents of social survey for restorability analysis, our research approach adopted)

IUCN’s criteria for ensuring the practicality and feasibility of the MPA	Contents of social survey for restorability analysis
	Item to be grasped	Questions
Compatibility with existing uses, particularly by locals	Level of human understanding of corals (High^a)	• Recognition of corals inhabiting the sea area respondents usually use• Benefit derived from corals• Recognition of the state of the coral
Social and political acceptability, i.e. degree of community support	Level of human motivation for coral conservation (High^a)	• Recognition for importance of coral conservation• Participation in coral conservation activities
Degree of insulation from external destructive influences	Level of destructive CES activities (High^a)	• Details of coral-related activities
Ease of management or compatibility with existing management regimes	Failure risk of offsetting (uncontrollable and local factors of coral degradation) (Low^a)	• Factors that lead to coral degradation
Accessibility for education, tourism, recreation	No corresponding survey items(Accessibility is precondition of candidate site selection)

^aIndicates preferable condition to secure the restorability.

Level of human understanding of corals

The level of human understanding of corals is estimated using the summation of respondent recognition levels of a coral inhabit, perception levels regarding the benefits derived from corals, and levels of recognition of the state of the coral. Each level is calculated by dividing the number of relevant answers by the number of valid answers. Regarding the recognition level of the state of the coral, the ratio of ‘Bad’ and ‘Very Bad’ answers to the actual damage to the coral is calculated to ensure the accuracy of the recognition level of the respondent.

Level of human motivation for coral conservation

The level of human motivation for coral conservation is estimated using the summation of the recognition level for the importance of coral conservation and the participation level for coral conservation activities. Each level can be calculated by dividing the number of relevant answers by the number of valid answers.

Level of destructive CES activities

Table 3 lists examples of options given for answering to the question required to understand the level of destructive CES activities. The correlation between the magnitude of engagement in the detailed activities (corresponds to options given in Table 3) and the dead coral cover should first be calculated from the answers using the formula given below, where Da is the number of respondents who engage in each detailed activity, V is the number of valid answers, Fx is the frequency of answers with respect to engagement in each activity (fishing, recreation, tourism business, and sea-sand mining) per valid answer, Px is the frequency of answers with respect to engagement in each activity per valid answer, Po is the population of an area or ward to which the target community belongs, and dc% is the percentage of dead coral cover. The destructive CES activities to be regulated at the offset site can then be identified through this correlation analysis.

Table 3. Example of options given for answering to the question required to understand the level of destructive CES activities

Questions	Options given for answering question
Methods and gears used for fisheries	Line and hook
	Net
	Trap
	Boat
	Spear
	Night fishing
	Anchor
Contents of recreation activity	Swimming
	Diving
	Relaxing
	Fishing
	Snorkeling
Contents of tourism business	Fishing service
	Eco tour
	Transportation service
	Diving/snorkeling service
Engagement in sea-sand mining	Yes/No

Engagement rate of each detailed activity

$ED=Da/V$

Weighted sum of answered frequency of each social activity

$WsFx=10\sum _{K=1}^3\left(WkFx\right)$

Weighted sum of answered period of each social activity

$WsPx=5\sum _{K=0}^5\left(WkPx\right)$

Magnitude of engagement in detailed activities

$Mda=ED\times WsFx\times WsPx\times Po$

Correlation between the magnitude and coral damage

$Correlation=\frac{\sum \left(Mda-\overline{Mda}\right)\left(dc\mathrm{\%}-\overline{dc\mathrm{\%}}\right)}{\sqrt{\sum {\left(Mda-\overline{Mda}\right)}^2\sum {\left(dc\mathrm{\%}-\overline{dc\mathrm{\%}}\right)}^2}}$

Failure risk of offsetting

Jacob et al. (2018) emphasized that the primary issues of offset site selection are insufficient diagnosis of environmental conditions of the site, connectivity to existing highly functional zones, and other anthropic pressures. Respecting his emphasis and local knowledge about anthropic pressures (i.e. failure risk of offsetting), the factors that lead to coral degradation can be asked in the interview survey. The options for answering are the following: (1) pollution, (2) climate change, (3) rainwater discharge, (4) fishing, (5) tourism, (6) local recreational activities, (7) sea-sand mining, (8) natural phenomenon, and (9) I don’t know. Among the options listed, 1 and 3 are the critical uncontrollable factors that ‘locally’ cause degradation. Therefore, the ratio of choices 1 and 3 to the total number of valid answers is calculated for each site.

2.3.3 Dependency analysis

To estimate human dependency on CES around candidate sites, the primary CES activities (e.g. fishery, recreation, tourism business, and sea-sand mining) should be identified through a pre-survey. Identifying the primary CES activities that relate to provisioning and cultural services is preferred in the light of the reality of recovery through offset schemes. Engagement in each CES activity, frequency, and period is asked in the interview. Regarding frequency, choices, such as 1–10 days, 11–days, and 21–30 days per month, are to be given to simplify the answering process and to avoid unit misunderstandings. Similarly, several activity periods of, for example, less than 1 year, 1–5 years, 6–10 years, 11–15 years, 16–20 years, and more than 20 years, are to be given. Engagement rate, weighted sum of engagement frequency, and engagement period are to be multiplied for each activity to optimize the engagement level. The formulas below were proposed to calculate human dependency, where E is the number of engagements per valid answer, f is fishing, r is recreational activities, t is tourism, s is sea-sand mining, W is given weight, F is the number of engagement frequencies per valid answer, and P is the number of engagement periods per valid answer.

Total engagement

$TE=Ef+Er+Et+Es$

Weighted sum of engagement frequency

$WsF=10\sum _{\begin{array}{c}k=1\\ x=1\end{array}}^3\left(WkFx\right)$

Weighted sum of engagement period

$WsP=5\sum _{\begin{array}{c}k=0\\ x=1\end{array}}^5\left(WkPx\right)$

Human dependency

$HD=TE\times WsF\times WsP$

2.4 Application to the project

2.4.1 Targeted project

The approaches described so far were applied to an actual coral offsetting case in Vanuatu to confirm their practicability. In this case, coral offsetting was required for a yen loan for the ‘Lapetasi International Wharf Development Project,’ a collaboration between the Republic of Vanuatu and the Japan International Cooperation Agency (JICA); the application story is described in the introduction of this paper. The original impacts of this project can be found in the related EIA report (EcoStrategic Consultant 2010). The offsetting principally requires securing additionality, meaning that the offset must provide conservation gains in addition to the planned or predicted conservation actions being taken by other parties (BBOP (Business and Biodiversity Offsets Programme) 2012b; Ledec and Johnson 2016a). In the case of Vanuatu, offsetting will be integrated in planned conservation action called CCA management, and restriction of activities will be newly set specifically for the offset; this specific restriction potentially secures additionality. However, measurements of additionality and conservation gains will inevitably be required (BBOP (Business and Biodiversity Offsets Programme) 2012a; Ledec and Johnson 2016). In the case of Vanuatu, measurements can be performed using data on the available area of the offset site, the annual growth rate of the corals at the offset site, and the rate of annual coral deterioration at the control site; detailed methods for measurement are omitted here, as they are beyond the scope of this paper. Additionally, the methodology to measure lost and restored CES should also be considered to secure social additionality and gains. Although the measurement is beyond the scope of this study, the changes of CES usage can be examined with the monitoring program. Thus, the level of engagement in CES activities could be a baseline and potential indicator of the monitoring.

2.4.2 Candidate site selection

In October 2017, a rapid-assessment was performed at 20 sites in the coastal area and islands in the Port Vila Bay of Vanuatu. Each site was evaluated using the criteria specified in Table 1. Accordingly, four candidate sites were selected: (1) Ifira East, (2) Ifira West, (3) Iririki East, and (4) Fatumaru (Figure 3).

Figure 3. Location of Port–Vila Bay, assessed sites (yellow pin), selected candidate sites (blue circles) and affected site (red squarish circle)

2.4.3 Data collection

Biological data collection

In October 2017, underwater observations and surface water measurements were conducted at candidate sites to understand the biological state. To evaluate the equivalency, data from the affected sites were collected by reviewing the report of project EIA conducted between October and November 2011. To ensure uniformity of the data quality, we respected the method and items of data collection previously used in the project EIA. A 2 m² quadrat was placed on the seafloor and used to collect data on dead and live coral cover, the number of coral species, the dominant coral, the number of benthos and seaweed/algae species, and substrate type. A data logger was used to measure water temperature, and salinity. The depth was measured by a dive computer.

Social data collection

In January and May 2018, structured interviews were conducted in communities that typically use the candidate sites, to evaluate restorability and human dependency and to identify destructive social activity. To compare the level and contents of CES lost at the project (affected) site with those of CES use at the candidate sites, the interview was also conducted in a community directly surrounding the affected site. The comparison was made to examine the variety of social characteristics in terms of CES usage and to set targets on the social state of offset achievement. The information from interviewed sites is described in Table 4. Because the biological survey revealed that Ifira West has minimal coral damage and would not deliver sufficient conservation gains, Ifira West was excluded from the interview target. The interviewees were Vanuatuans between the ages of 11 and 79; gender, affiliation, religion, and tribe were not considered. To avoid any response bias from interviewee expectations of the project, incentives such as gifts and/or money were not given. Additionally, the interviews were performed individually to avoid response bias resulting from communication among interviewees.

Table 4. Information of interviewed sites

Site name	Name of interviewed community	Name of area/ward, the community belongs	Population of the area/ward	Number of interviewees	Gender ratio of the interviewees (Male/Female)
Ifira East	Ifira	Ifira Island	1,186	90	55/35
Iririki East	Seaside paama/tongua	Cental Ward	4,842	100	54/46
Fatumaru	Fresh wind/Numburu	Anabrou-Melcoffee Ward	10,123	100	61/39
Affected site	Ifira	Ifira Island	1,186	91	58/33

3. Results

3.1 Restorability

3.1.1 Biological state

The biological states of the affected and candidate sites are shown in Table 5. The dominant substrate type was sand, but coral rubble and rock were also present at Ifira East and Ifira West. The depth ranged from 1.3 to 2.1 m, except for 4.4 m at Ifira West. The water temperature was similar for all the sites but slightly lower (26.8°C) at the affected site and higher (28.1°C) at Ifira East. The salinities of the affected sites, Ifira West, and Ifira East showed relatively high values of 34.8‰, 33.3‰, and 33.0‰, respectively, contrasting with those of Iririki East and Fatumaru (30.9‰ and 29.6‰, respectively). This difference is likely because of the physical distance between the former three sites and the latter two sites. The affected site and Fatumaru showed similar levels of live coral cover (10.7% and 10.0%, respectively), whereas the dead coral cover was very different (0.2% and 60%, respectively). Ifira East had the closest amount of overall coral cover – expressed as the sum of dead and living coral cover – to the affected site (15.0 and 10.9, respectively). Ifira West had the average number of coral species closest to that of the affected site (5.0 and 4.3, respectively), followed by Ifira East (6.0). Ifira East and West had similar Shannon diversity index (H’) values (2.4 and 2.3, respectively), and Iririki East and Fatumaru had the same value (0.4), likely because of the physical distance between these sets of sites. Ifira West had the number of benthic invertebrate species closest to that of the affected site (3.0 and 3.2, respectively). However, all the sites, except Ifira West (1.0), had similar values (ranging from 3.0 to 3.2) for the number of seaweed and algae species.

Table 5. Biological state of the affected site and offset candidate sites

Parameters		Affected site (Average of 13 sub-sites)		Candidate sites
				Ifira East	Ifira West	Iririki East	Fatumaru
Physicochemical environment	Substrate type	Sand		Coral rubble and sand	Rock and sand	Sand	Sand
	Depth (m)	2.1		1.3	4.4	1.7	1.3
	Temp. (°C)	26.8^a		28.1	27.4	27.2	27.2
	Salinity	34.8^a		33.3	33	30.9	29.6
Coral	Dominant	Psammocora sp. /Porites sp.		Astreopora gracilis	Astreopora myrioohthalma	Acropora aspera	Prorites australiensis
	Live coral cover (%)	10.7		5.0	25.0	50.0	10.0
	Dead coral cover (%)	0.2		10.0	0.0	15.0	60.0
	Coral cover (%)	10.9		15.0	25.0	65.0	70.0
	Number of species	4.3		6.0	5.0	2.0	2.0
	Shannon diversity index (H’)	1.3		2.4	2.3	0.4	0.4
Benthos	Dominant	Diadema savignyi		Echinometra mathaei	Diadema savignyi	Holothuria atra Demospngeae	Propeamussiidae
	Number of species	3.2		2.0	3.0	2.0	5.0
Seaweed /Sea alga	Dominant	Peyssonneliaceae		Halimeda sp. Jania sp. Boodlea coacta	Melobesioideae	Halimeda sp.	Boodlea coacta
	Number of species	3.2		3.0	1.0	3.0	3.0

^aData were collected at single sub-site

Figure 4 shows the results of the cluster analysis. A cluster formed between Ifira West and Ifira East, and another cluster subsequently formed with the affected site at a level of almost 87%. An additional cluster was formed between Iririki East and Fatumaru; this then formed another cluster with the remaining three sites at a level of roughly 77%. Therefore, the biological states of Ifira West and Ifira East are more like that of the affected site.

Figure 4. Results of cluster analysis on biological state

3.1.2 Social perspectives

Table 6 shows the standardized data for the restorability analysis; bio-equivalency was derived from cluster analysis of biological data (Figure 4). Interpretation of the data is described below.

Table 6. Standardized data for restorability analysis

Candidate site		Bio-equivalency(High^a)	Level of human understanding of corals (High^a)	Level of human motivation for coral conservation (High^a)	Level of destructive CES activities (High^a)	Failure risk of offsetting (uncontrolable and local factors of coral degradation) (Low^a)
Ifira East		1.4	0.7	−0.7	−0.8	−1.3
Iririki East		−0.7	−1.4	1.4	−0.7	1.2
Fatumaru		−0.7	0.7	−0.8	1.4	0.1

^aand underline (_) indicate preferable condition.

Level of human understanding of corals

Human understanding of corals was high in Ifira East (0.7) and Fatumaru (0.7). At those sites, consensus for offsetting is likely easily obtained; therefore, the management of coral resources is supposedly promoted. At Ifira East, higher human dependency mentioned later may lead to a higher understanding of corals. In contrast, human understanding was low in Iririki East (−1.4), maybe because the communities interviewed for Iririki East are located in the nearby commercial area of Port Vila city that features an urbanized lifestyle and relatively lower interest in and understanding for coral resources.

Level of human motivation for coral conservation

Human motivation for coral conservation was relatively higher in Iririki East (1.4) than in Ifira East (−0.7) and Fatumaru (−0.8). In the communities interviewed for Iririki East, residents supposedly had frequent opportunities to participate in the conservation program, perhaps leading to higher motivation. A low understanding of corals combined with a high motivation for coral conservation in Iririki East may indicate an imprinting of the importance of environmental protection without a basic knowledge of ecology. Education to promote understanding of coral should be given if Iririki East is selected as an offset site.

Level of destructive CES activities

The level of destructive CES activities was relatively higher in Fatumaru (1.4) than in Ifira East (−0.8) and Iririki East (−0.7) and may be related to the fact that the Fatumaru had the highest population of communities interviewed (Table 4). If Fatumaru is selected as an offset site and is well managed, the damaged corals can be expected to efficiently recover.

Failure risk of offsetting

The risk of failure of offsetting was relatively higher in Iririki East (1.2) than in Ifira East (−1.3) and Fatumaru (0.1). In Iririki East, water pollution is primary recognized as a factor that led to coral degradation, and this causal relation between polluted water and coral degradation is consistent with personally communicated information from the Secretariat of Pacific Regional Environment Program (SPREP). The Government of the Republic of Vanuatu implied that the probable cause of the high pollution in Port Vila Bay is the lack of a sewage system and poor management of multiple individual septic tank systems (personal communication). There is a risk that the degraded corals will not recover if Iririki East is selected as an offset site.

3.2 Dependency

In this paper, human dependency on CES was estimated by calculating levels of human engagement in CES activities. Although all investigated sites are located within a small area, the calculation obtained variation in the level of engagement (Figure 5). Engagement in the fishing and tourism business was almost equally high in Ifira East and the affected site, and engagement in recreation was significantly higher in Ifira East. Fishing and recreation retain a consistent portion of the engagement across the investigated sites. In contrast, the engagement in sea-sand mining was constantly low among the candidate sites. Thus, fishing and recreation are major local CES activities in the Port Vila Bay.

Figure 5. Engagement of locals in CES activities

The data collected for dependency analysis are integrated in Table 7. The lowest and highest levels of engagement in CES activities (TE) were found in Fatumaru (1.0) and Ifira East (2.1), respectively. In Fatumaru, local individuals may have less opportunity to perform such activities, as they can easily access the commercial area of Port Vila city for their livelihood. Conversely, in Ifira, people may rely more heavily on natural resources because of the long distance from a commercial area and because the community in Ifira has a customary right to manage marine resources of Port t Vila Bay. The lowest engagement frequency and period values were found in Iririki East (42.2 and 15.8, respectively), perhaps because of the short distance to a commercial area. Iririki Island, where Iririki East is located, is dominated by a single resort hotel, and few locals regularly use the coral resources around the Island. This may also be related to the lowest engagement frequency and period reported in Iririki East. Based on these, Ifira East and Iririki East have the highest (5,956) and lowest (848) degree of human dependency, respectively.

Table 7. Dependency on CES

		Total engagement in CES activities (TE)		Weighted sum of engagement frequency (WsF)		Weighted sum of engagement period (WsP)		Dependency(TE×WsF×WsP)
Ifira East		2.1		72.4		39.8		5,956
Iririki East		1.3		42.2		15.8		848
Fatumaru		1.0		65.5		20.9		1,342
Affected site		1.1		52.9		72.8		4,272

3.3 Candidate site evaluation

Figure 6 illustrates the advantages and disadvantages of the candidate site by plotting standardized restorability and human dependency data and expressing estimated accessibility by label size. Ifira East is positioned in an area with social risks. Because of its higher human dependency, individuals may oppose restrictions on CES use, conflict may occur because of the intensive use of resources after biodiversity gains are achieved, or additional CES usage losses may occur because of restrictions on relevant activities for offset site management. Iririki East is positioned in an area with failure risks. Because of its lower restorability, sufficient cooperation may not be obtained from local CES users for offset site management; Iririki East may also be influenced by uncontrollable factors such as the runoff of untreated water from commercial land, which may negatively affect local coral growth. In contrast, Fatumaru is positioned in an advantageous area, where moderate restorability, human dependency, and accessibility coexist. In such an area, the efficient recovery of coral ecosystems and CES is expected, and the negative impacts of the offsetting on CES activities are comparatively smaller.

Figure 6. Advantages and disadvantages of candidate sites. Size of label circle indicates an advantage in accessibility (a larger size means greater accessibility)

3.4 Activities to be regulated at the offset site

The details of CES activities that show correlation with dead coral cover have been identified (Table 8). The results indicate that fishing with a line, hook, or spear has a positive correlation with mechanical coral damage. Use of this fishing gear often involves individuals walking in a shallow area, which likely causes mechanical damage. When spear fishing, individuals often stand on the coral in a shallow area to rest and catch their breath, thus causing damage from trampling; recreational activities did not exhibit any correlation. In terms of tourism, the ecotour business showed a positive correlation. Although the ecotour business itself is unlikely to have an impact on coral, some damage may occur if proper guidance is not given. Overall, it is worth noting that a population engaging in these activities is likely to affecting the correlation. Moreover, most of the activities, except relaxation, require direct passage through shallow areas of coral. Therefore, most of the activities have the potential to be restricted for offset site management.

Table 8. Details of CES activities and correlation with dead coral cover

Details of coral-related activity		Correlations with dead coral cover %
Fishing methods and gears used	Line and hook	0.9985	*
	Net	0.9090
	Trap	−0.5766
	Boat	0.5570
	Spear	0.9975	*
	Night fishing	0.8828
	Anchor	N/A
Contents of recreation activities engaged	Swimming	0.9955
	Diving	0.9232
	Relaxing	0.9944
	Fishing	0.9280
	Snorkeling	−0.0285
Contents of tourism business engaged	Fishing service	0.9449
	Eco-tour	0.9981	*
	Transporatation service	N/A
	Diving/sonorlkeling service	N/A
Engagement in sand mining	Sand mining	0.8268

*Strong correlation at 5% significance level.

N/A: No answering or less answering.

4. Discussions

4.1 Restorability

Figure 7 illustrates the advantage of considering restorability. With conventional offsetting, the coral ecosystem can be restored to its original state (the state of the affected area), but fully restoring CES usage is difficult. However, if restorability is considered, the chances of restoring CES usage increase. In the case of Vanuatu, Fatumaru exhibits only moderate restorability, partially because of low motivation for coral conservation. Therefore, if motivation is improved through measures such as environmental education, restorability (e.g. coral growth) will increase, and restoration of CES usage will be more feasible. In sites with higher CES usage, such as Ifira East, other types of restorability enhancement, such as awareness creation for the sharing of restored CES, might facilitate usage without conflict. Both cases should be accompanied with environment-friendly usage to avoid further coral damage. Thus, the restorability consideration approach promotes the mitigation of the social issues of offsetting, such as additional loss of ES and/or conflicts because of intensive ES use.

Figure 7 Advantage of restorability consideration

4.2 Dependency on CES

Application of the approach revealed that, despite a relatively small spatial extent, there is a large difference in human dependency on CES among offset candidate sites. Such differences are presumed to be related to the distances between the communities using these sites and the commercial area. A community located closer to the commercial area is more exposed to modern products and cultures, consequently, less dependent on natural resources. The results from communities in which Iririki East users live reflect the location near a commercial area and the low human dependency on CES. The other potential factor that regulates such a difference is the presence of a customary resource management system. In Vanuatu, the Constitution recognizes the right of customary owners to control their own land (Tacconi and Bennett 1995). This system is also applicable to marine resources (author’s personal communication with government officer of Vanuatu). The marine resources of Port Vila Bay are fundamentally managed by a single community, the so-called ‘Ifira.’ The high human dependency on CES and low potential for coral damage in Ifira East likely indicates the effectiveness of this resource management system.

Griffiths et al. (2019) noted that the importance of considering localized use and no-use values (such as cultural value) of ES in offset practices change depending on geographic conditions. In our approach, the localized use and non-use values of ES are made clear via dependency analysis, implying that our approach functions well. From a more detailed social perspective, the inextricable link and complex relationship between human and natural resources is widely recognized (Holling 2001; Levin 2006; Brownlie et al. 2013) and is even more pronounced in developing countries where geographic conditions vary within a small area. McClanahan et al. (2006) suggest that, in terms of corals, considering local socioeconomic factors is essential to improving the ability of coral reef management regimes. The dependency analysis of our approach confirms, practically, that a deeper understanding of such social and environmental complexity is a fundamental social consideration for the management of coral reefs in developing countries.

Dependency analysis contributes to the consideration of vulnerable individuals. In the safeguard policy of international aid organizations, there is a general principle that special consideration should be given to the most vulnerable individuals (Bidaud et al. 2017). Vulnerable individuals are defined as those who are less likely to benefit from a project (The World Bank 2004). Thus, individuals highly dependent on ecosystems can be interpreted as vulnerable, because it is often difficult for them to directly benefit from development projects because of their physical distance from the project site. Our approach provides information useful for avoiding the selection of offset sites with higher human dependency; this mitigates the impact on vulnerable individuals.

Although the selection of offset sites with higher human dependency cannot mitigate social issues, those sites still have a chance to be an offset site if effective supplemental measures are taken. For example, Ifira East showed the highest human dependency, especially in fishing and recreation; this situation may cause conflict because of the intensive use of resources after the achievement of biological gain. However, if supplemental measures such as the provision of alternative sites for these activities or of alternative means of livelihood are proposed, Ifira East has the potential to be an offset site. Additionally, environment-friendly measures, such as the improved use of CES, dispersion of usage, and environmental awareness should be accompanied with supplemental measures to avoid further resource degradation.

4.3 Restriction of social activities

Screening of the restriction of destructive CES activities could be effective in successful offset and sustainable coral resource management and ultimately for the creation of win–win relationships. In this study, potential destructive CES activities were identified by interviewing locals and examining the correlation between detailed CES activities and dead coral cover. These outputs can be used for developing an offset site management plan, although further on-site observation is required to confirm this correlation. The restriction of CES activities may negatively impact the livelihood of the users; therefore, providing effective support for restoring livelihood or (preferably) enabling continuous engagement in CES activities at alternative locations is important.

5. Conclusions

Selection of a site with higher restorability, lower human dependency on CES, and selective restriction of social activity are vital considerations approach to mitigate social issues that may arise after the selection and establishment of an offset site. Furthermore, the understanding of social characteristics and their relationship with CES is the fundamental base for applying these consideration approaches in coral offset planning. We tested these approaches on the coral offsetting case in Vanuatu and determined that the aforementioned characteristics and relationships vary among candidate sites, despite all sites being located in a small area. Furthermore, we revealed that each candidate site had advantages and disadvantages in terms of restorability, human dependency on ES, and accessibility. These spatially diversified characteristics are substantially common in developing countries. Therefore, comprehensively evaluating candidate sites to identify optimal offset sites is desirable. Candidate sites with disadvantages still have the potential to become offset sites if supplemental measures are taken. However, completely negating all issues is difficult, because the proposed approaches and supplemental measures are primarily concerned with mitigation. Hence, the remaining issues should be complemented by the provision of compensation with environment-friendly measures to effectively achieve offsetting.

Offsetting has conventionally focused on biological validity; we propose a novel approach that incorporates social elements into offsetting. This approach, by considering social aspects, provides several suggestions for implementing offsetting in developing countries where people tend to rely on natural resources. From a coral resource management perspective, this approach facilitates both conservation and the sustainable use of coral resources by mitigating social issues related to CES; therefore, the tradeoff that exists between coral conservation and usage can evolve into a win–win relationship. Our approach improves the regulatory service (e.g. flood control) and provisioning services (e.g. food security) of corals by restorability consideration and sustainable resource management perspectives, respectively, and thereby can potentially contribute to climate change adaptation.

This study does not discuss the participatory approaches in depth, mutual agreement, and cooperation among stakeholders, including local people, that are necessary for the successful restoration of corals and CES. Therefore, coordination among the stakeholders to facilitate the restoration should be performed in parallel to the application of the approach suggested in this study.

Acknowledgments

The authors thank all staff members at the Vanuatu Project Management Unit, the executing agency of the Lapetasi International Wharf Development Project, for their support during data collection; the authors thank staff members at the Department of Environment Protection and Conservation for their information about the Community Conservation Area.

Disclosure statement

No potential conflict of interest is reported by the authors. Although the research cited throughout this paper was planned as part of the PhD studies undertaken by the lead author, the interviews were conducted by the Japan International Cooperation Agency (JICA), a donor agency of the wharf development project. Therefore, all data shown in this paper were provided by JICA. The views expressed in this paper, however, are those of the authors alone and do not reflect the official view of JICA. The author did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

Notes

1. Restorability indicates the possibility of restoring ecosystem services through ecosystem restoration.

2. In this study, level of engagement is used as a term that indicates the degree to which a person engages in a particular activity.