Tài liệu Dịch vụ sinh thái rừng ngập mặn và nuôi tôm

Ecological Indicators 46 (2014) 201–213 Contents lists available at ScienceDirect Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind Wetland ecosystem service values and shrimp aquaculture relationships in Can Gio, Vietnam S. McDonough a, * ,1, W. Gallardo b , H. Berg c, N.V. Trai d , N.Q. Yen d a Aquaculture and Aquatic Resources Management, Asian Institute of Technology, Klong Luang, Pathumthani, 12120 Bangkok, Thailand Department of Marine Science and Fisheries, College of Agricultural and Marine Sciences, Sultan Qaboos University, Oman c Department of Physical Geography & Quaternary Geology, Stockholm University, Sweden d Department of Fisheries Management and Development, Faculty of Fisheries, Nong Lam University, Linh Trung Ward, Thu Duc District, Ho Chi Minh City, Viet Nam b A R T I C L E I N F O A B S T R A C T Article history: Received 2 February 2014 Received in revised form 9 May 2014 Accepted 10 June 2014 Wetland valuation methods often apply monetary driven approaches that may undermine intrinsic ecosystem values. Utilizing a stated preference method, the study identified and mapped local stakeholder ecosystem service values between subsistence wetland and shrimp farmer groups in Can Gio, Vietnam. Through focus group choice experiments, ecosystem services correlating to unique price increments and cost/benefit tradeoffs between hypothetical intensive aquaculture developments and mangrove conservation scenarios were investigated. Selection outcomes exhibited strong values for ecosystem services maintained at the hypothetical natural state (core area pre-intervention 45%, postintervention 55%; buffer zone pre-intervention 65%, post-intervention 73%). Few respondents selected the hypothetical intensive aquaculture development scenario (core area pre-intervention 18%, post-intervention 9%; buffer zone pre-intervention 12%, post-intervention 5%), and instead, most respondents avoided the costs and benefits of intensive aquaculture development in preference for maintaining natural ecosystem services. Group deliberations drew out a higher sense of altruism and responsiveness to intrinsic wetland values that superseded the potential economic gains of aquaculture developments, whereby certain ecosystem services were deemed economically unassociable and irreplaceable for both study groups. The qualitative results expose the difficulties in monetarily measuring ecosystem services, highlighting the need to incorporate approaches that integrate the intrinsic values attached to ecosystem services. ã 2014 Elsevier Ltd. All rights reserved. Keywords: Mangroves Ecosystem services Environmental values Aquaculture Vietnam 1. Introduction Ever since (post Vietnam war economic reforms), Vietnam’s wetlands have been, and are still subject to conversion for monetary gains driven primarily by high export demands for shrimp. Vietnam’s growing shrimp industry has raised environmental, economic and social attentions, and despite numerous restoration projects, high mangrove losses as a result of aquaculture development have been reported (World Bank, 1999; World Atlas of Mangroves, 2010). In 1976, Vietnam produced around * Corresponding author at: Aquaculture and Aquatic Resources Management, Asian Institute of Technology, Klong Luang, Pathumthani, Bangkok 12120, Thailand. E-mail addresses: [email protected] (S. McDonough), [email protected] (W. Gallardo), [email protected] (H. Berg), [email protected] (N.V. Trai), [email protected] (N.Q. Yen). 1 Permanent address: FAORAF, Gamel Abdul Nasser Road, P.O. Box 1628, Accra, Ghana. Tel.: +233 505256401. http://dx.doi.org/10.1016/j.ecolind.2014.06.012 1470-160X/ ã 2014 Elsevier Ltd. All rights reserved. 200,000 t of shrimp and in 2011, a significant production increased with over 487,000 t covering over 290,000 ha of wetlands was recorded on a national level (Wilder and Phuong, 2002; EJF, 2003; FAO, 2011). Unregulated wetland conversion into shrimp farms combined with detrimental aquaculture practices influences natural habitat resiliency and its ability to deliver supporting, regulating, provisioning and cultural ecosystem services defined by the United Nation’s Millennium Ecosystem Assessment (MEA, 2005; TEEB, 2010) and exposes stakeholders to different natural and socio-economic risks. The true long-term outcomes of shrimp aquaculture development costs and benefits are not fully understood with shrimp aquaculture development potentially more costly through its long-term externality costs (Primavera, 1997; Ashton, 2007), than the preservation of wetlands and its delivery of natural ecosystem services to local and regional communities, directly and indirectly. Shrimp aquaculture inputs and outputs are economically driven and exploitative with little considerations to the long-term 202 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 negative environmental outcomes (Rosenberry, 1998; Páez-Osuna, 2001; EJF, 2003). However, the economic benefits of shrimp aquaculture should not be disregarded as added capital can boost the national economy, improve local livelihoods, and aid in rural development (Frankic and Hershner, 2003; Gunawardena and Rowan, 2005). Vietnam’s aquaculture is poised to intensify as declared by Government Decision No. 21, 1998 and Government Decision No. 67, 1999 (EJF, 2003), encouraging aquaculture expansion at the private level, exposing mangrove wetlands to future pressures. Wetland habitats should not be overlooked for a single industry as they also support other sectors such as tourism, fisheries and sustain rural livelihoods by providing provisioning services such as plant and animal materials used for housing, medicines, foods, and fuel (Rönnbäck, 1999). Unsustainable shrimp developments combined with new challenges such as climate change and hydropower developments (Hori, 1993; Bakker, 1999) presents the ecosystem service concept as a tangible approach to utilize in habitat valuation to better mainstream a more ‘risk-proof’ approach in policy development when managing land-altering industries, such as aquaculture to be a more proactive industry rather than its typical reactive manner. Shrimp developments, particularly in Southern Vietnam is often focus on intense production and are detrimental to the delivery of ecosystem services (World Atlas of Mangroves, 2010). Can Gio presented a unique location based on its close proximity to Ho Chi Minh City and the UNESCO biosphere reserve to examine the exchanges in conservation and development values. As the core area is protected there is no imminent threat to the wetland, but intensification and future expansion of shrimp farming activity from buffer zone activity will have impacts on habitats. Utilizing ecosystem services as the core component of the choice experiment, the study identified and mapped what values were placed on wetland ecosystem services in relation to natural conservation and shrimp aquaculture developments in Can Gio’s Fig. 1. Location of core area and buffer zones where focus group discussions were conducted. Shrimp farming is undertaken within buffer zones under the respective districts. Core area focus groups were conducted in the central areas of Can Gio. S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 socio-economic and biophysical context. Choice experiment focus groups were conducted in two respondent groups: subsistence wetland users living inside the UNESCO biosphere reserve core area and shrimp farmers along the reserve’s buffer zone. Focus groups were utilized as a means for the respondents to discuss and declare what ecosystem services they valued individually and as a group under the different hypothetical scenarios presented. The study objectives were to (1) identify the principal stakeholders reliant on mangrove wetland ecosystem services, (2) identify what ecosystem services are present, delivered, sustained, and utilized by these wetland users, (3) identify if and how shrimp aquaculture developments or wetland protection impact ecosystem service delivery, (4) map the communal interactions and values placed on mangrove ecosystem services in relation to wetland conservation and aquaculture development drivers, and (5) assess what influences trade-offs and degrees of willingness to accept price increments between the different ecosystem service scenarios amongst respondent groups. 2. Material and methods 2.1. Study area and groups Rich in flora and fauna biodiversity, the UNESCO Can Gio biosphere hosts over 21,000 ha of wetlands with 77 different mangrove species (UNESCO, 2000). Can Gio’s proximity to Ho Chi Minh City and shrimp farming activity around the buffer zones make it an ideal location to examine the anthropogenic values and its relationship to ecosystem services, shrimp aquaculture development, and wetland conservation. Shrimp farming in Can Gio occurs in the transitional buffer zones around the UNESCO reserve where in 2012, four districts were recorded to regularly conduct shrimp farming (Fig. 1). An Thoi Dong ward hosts around 695 operating shrimp farms; Binh Khah, 508; Ly Nhon, 243 and Tam Thon Hiep, 68 shrimp farms; in total, Can Gio encompasses around 1514 officially recorded operational shrimp farms (Can Gio Annual Report, 2011). Key informants from the Can Gio Peoples’ Committee, the Can Gio Mangrove Protection Management Board Park Authority, and Nong Lam University in Ho Chi Minh City assisted in identifying respondent groups, providing logistic, linguistic and personnel support in both core areas and buffer Fig. 203 zones. Core area respondent groups who mainly live a subsistence lifestyle within mangrove habitats were investigated to tease-out what ecosystem services are valued in a more naturally unperturbed situation with minimal aquaculture developmental influences. Core area groups consisted of three stakeholders groups: artisanal fishers, intertidal small-scale mollusc farmers, and park rangers. Artisanal fishers comprised of individuals that lived on boats, fishing Can Gio’s tributaries. Most mollusc farmers lived in houses constructed on intertidal mudflats and cultured mollucs species such as Crassostrea gigas, Anadara granosa, and Meretrix lyrata inside the reserve’s core areas. Park rangers consisted of rangers employed under the Can Gio Mangrove Protection Management Board Park Authority and lived in posts situated throughout Can Gio’s core areas. Differences between intensive and extensive shrimp farming were not made as buffer zone groups were randomly conducted. The total numbers of respondents within the core areas were lower in number due to time, boating costs and personnel constraints. In total, 157 individuals participated in the choice experiment, 42 questionnaires distributed with 10 focus groups consisting of 42 participants from the core area (16 fisher folk, 8 mollusc farmers and 16 park rangers) and 32 focus groups of 118 participants (mostly shrimp farmers) in the buffer zones. 2.2. Research design Stated preference methods have been applied throughout the region to investigate wetland values in Do and Bennett (2005) and Thuy (2006). In this study, a series of focus group discussion choice experiments identified what ecosystem service utility values existed between the two stakeholders and distinguished the degree of communal willingness to accept price increments between the costs/benefits of shrimp development and mangrove conservation efforts. The stated preference technique, rather than the revealed preference technique, was applied as it minimized biases and allowed for the presentation of hypothetical scenarios, enabling greater content flexibility and comparisons (Hanley and Spash, 2001) between the studied groups. Focus group choice experiments (Fig. 2) began with a warm-up exercise whereby three items related to wetland habitats (Nypa fructans seed nodule, Pluchea indicus herbs and insect repellents) were presented to 2. Steps taken in the focus group activity to deliberate ecosystem service values in relation to shrimp aquaculture development and wetland protection. 204 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 stimulate group discussions on habitat alteration. Warm up discussions did not contribute to the final datasets and lasted around fifteen minutes with discussions taking place on item usage, abundance and frequency of its access. Next, a specific choice set was presented to the particular respondent group beginning the deliberation process. After choices were selected and discussed, the intervention exercise would be conducted to further probe respondent ecosystem service values in relation to shrimp aquaculture and wetland conservation. After the intervention exercise, the same choice set was presented to the same respondents to once again deliberate and discuss the choice selections, questioning the changes, if any were made. At the end, a round up discussion was conducted debating and summarizing what has been learned from the discussions. One of the most important aspects of this approach was the empowerment of local communities and the opportunity for the respondents to choose the desired ecosystem service outcomes based on their experiences. The respondents’ opportunity to discuss the reasons behind their choices made before (pre-intervention) and after (postintervention) the intervention exercise helped to elicit deeper discussions on what communal values were placed on their habitat (see Section 2.3). 2.2.1. Ecosystem service scenarios and price increments The stated preference technique required respondents to determine their most preferred option from the three different scenarios presented as ‘states’. These states characterized ecosystem services delivered in bundles at different proportions (Fig. 3). The natural state is representative of an unperturbed habitat with minimal aquaculture developments due to higher conservation efforts. The current state attempts to best represent the local situation at that specific time for that specific site; a ‘middle ground’ that does not reflect intensive aquaculture development or strict wetland protection. The alternative state reflects a landscape shaped by human enterprise and engineering altered for aquaculture production at higher intensities resulting in altered habitats for economic outcomes. For the sake of the experiment, states were not revealed during focus groups discussion until the later stages of the task (during post-intervention choice selections), but were presented as price increments values, pre-intervention. Price increments were placed on each state with the aim to tease out ecosystem service values in relation to monetary costs when the services are faced with shrimp aquaculture development pressures or wetland conservation pressures between the two study groups. Within these hypothetical states different services and products would be delivered, some at higher or lower rates and at different prices depending on the state and the ecosystem service (see Section 2.3). Fig. Core area price increments were presented as a hypothetical government wetland conservation plan whereby communities paid an annual community tax. Prior to the choice experiment it was revealed, through key informant interviews that a 7,000,000 VND yearly tax is placed on inhabitants living inside core areas and this value was used as the current state median value between the alterative and the natural state (Table 1). The alternative state declares a hypothetical 25% increase in the communal tax is required to manage shrimp farming impacts (e.g., clean up schemes and personnel to regulate developments and practices) placing core area alternative price increments at 8,750,000 VND annually. The natural state declares that hypothetically higher taxes are placed to maintain more stringent conservation efforts (ranger fees, station construction, maintaining laws, and legislations) therefore price increments increases 50% (10,500,000 VND annually) and as a result, natural ecosystem services are delivered more efficiently. Buffer zone price increments were presented as shrimp farming production costs such as feed, anti/probiotics, electricity, labor, and other production expenses. The current state assumes a median base where production costs are contextual to the period of the study when discussing with key informants and pilot study shrimp farmers where it was revealed that most Can Gio farmers spent around 190,000,000 VND/crop. Based on this figure shrimp farming price increments were presented as either a marginal 25% or a 50% increase in production costs relative to the specific respondent farmers. Consequently the current state was presented as a “no change” in farming costs state, the alternative state assumed that higher shrimp production outputs and inputs resulted in higher costs (50% increase in production costs) with the consequence of ecosystem services delivered at a degraded rate. However, the benefit of higher production per cycle may result in greater monetary profits were presented as a possible outcome. The natural state presents lower production, but due to a lack of infrastructure (electricity, roads, and feed outlets) the costs to acquire the necessary farming products and services are hypothetically less accessible resulting in a 25% increase in overall costs. It was stressed to participants that although these scenarios were hypothetical, respondents should try to treat it as real life outcomes. To help devise scenario parameters and gain an understanding of the current situation, key informant interviews with Can Gio People’s Committee and Park Authorities were conducted before the actual focus groups. After the key informant interviews, a preliminary field visit to study locations was conducted to carry out two pilot studies, one each for core area and buffer zone groups. The pilot study lasted two days and enabled for the adjustment and improvement of the study methods, helping refine the hypothetical states presented, tested 3. The three states and their ‘bundled’ efficiency conditions presented to the respondent groups. S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 205 Table 1 Price increment mechanism and effects under their unique states presented to respondent groups. Price increment Core area Natural 10,500,000 VND annual tax Mechanism Increased efforts on wetland conservation projects outside core area. Effects Increased delivery of natural ecosystem services and richer biodiversity. Decreased access to market goods, reduced local infrastructure and economy. Buffer zone Current 7,000,000 VND annual tax Alternative 8,750,000 VND annual tax Natural 25% increase in production costs No changes in conservation efforts. Core area boundaries remain the same. Decreased efforts on wetland protection inside core area. Private shrimp venture compensate specific costs. Low infrastructure results in higher prices, middleman services, electricity costs, price for fuel to access feed, and seed. Some ecosystem services delivered with moderate access to markets outside core area. Decrease delivery of natural ecosystem services, area less biodiverse with a degraded natural habitat. Improved infrastructure and introduction of urban market systems, better access to market goods. the initial demographic questionnaires, and scrutinizing the ecosystem service parameters. 2.3. Choice experiment focus groups and choice grids Focus groups discussion were conducted with mostly males, not age specific, with a minimum of three to four individuals and with discussions lasting one and a half hours. The initial task of each focus group discussion was to obtain individual demographic information (age, monthly salary, education, years of shrimp farming/living in reserve areas, if they possessed radios/TVs, and if they had any children). Discussions were recorded using an audio recorder for all the respondent groups and later translated from Vietnamese into English. Throughout the study, four different choice sets were prepared and presented to four different groups matching their context. One choice set for each study group was presented and explained by the study facilitator. The choice grids were fabricated from a plastic sheet with a grid drawn onto it with picture cards used to convey ecosystem service parameters as written text would exclude illiterate individuals, hold attentions better and was easier for the facilitator to explain. Images were placed on their respective position on the grid and coloured stones were used to trace respondent choices, pre and post-intervention. Four stones were given to each individual to place on the choice grid where each individual had to choose once per row (ecosystem services) and must consider the conditions of each of the three columns (state/price increment) presented. To sum up, each respondent had four choices out of a possible twelve; by avoiding or selecting a certain choice, respondents indicated their ecosystem services values and the desired outcomes of their environment and livelihood. The facilitator constantly reminded respondents to consider price increments and their consequences throughout the discussions, individually and as a group, promoting group discussion on the topics raised and asking for explanations behind the choices made and the choices not made. After selections were made and the reasons behind the selections discussed, the facilitator would then conduct the intervention exercise aimed to stimulate respondents to discuss comprehensively the potential impacts on ecosystem services driven by aquaculture development or wetland conservation efforts. To initiate the intervention exercise, the facilitator would Current No changes in production costs Shrimp farming methods and activity resume as regular. No changes in shrimp farming methods. Increased delivery of natural Some ecosystem ecosystem services, higher services delivered biodiversity and natural with moderate heritage. access to markets Decreased access to market outside core area. goods, reduced local infrastructure and economy. Alternative 50% increase in production costs Increased production intensity results in higher farming costs, labour, medicines, electricity, logistics, feed and seed costs are the drivers. Decrease delivery of natural ecosystem services, less biodiverse and degraded natural habitat. Improved infrastructure and introduction of urban market systems. Improved livelihood for successful farmers. ask the respondents if they knew examples of how shrimp aquaculture and conservation efforts alter landscapes and ecosystem services. If respondents were unable to provide examples and discussion was lacking, the facilitator would present a set of images, in random order, showcasing three negative (images of degraded mangrove habitat, shrimps infected with hepatopancreatic parvovirus, and urban pollution) and three positive examples (images of possible monetary profits, better access to services/ infrastructure and new opportunities for private enterprises) of how shrimp aquaculture and mangrove conservation projects can alter wetland habitats. Intervention discussions lasted around forty minutes where afterwards (post-intervention) the same choice grid was again presented to the same respondents for reselection with continued discussions on the reasons behind their change or lack of change. To conclude a focus group discussion, a round up session was conducted with respondents scrutinizing the issues and implications associated with shrimp development or wetland conservation on ecosystem services and their values placed on its access with the frequency of topic mention tallied during round up discussions. In addition to revealing the learning outcomes, a comprehension rating was given when focus group discussions terminated to rank the quality of respondent outcomes. This component is based on the respondent’s understanding of the choice grid, their selections, and the quality and amount of respondent feedback during both intervention exercise and round up discussions. Ratings ranged from 1 to 3; (1) being of average quality discussion, (2) good quality discussion, and (3) excellent quality discussion. Buffer zone shrimp farmers were presented one grid and core area respondents presented three choice grids for each respective artisanal fishers, mollusc farmers, and park ranger groups. Four constant parameters were presented throughout the study: direct provisioning, indirect provisioning, regulating, and cultural ecosystem services. Row one was presented as a direct provisioning good related to the livelihoods of the specific responding group; row two, as an indirect provisioning service; row three, regulating/habitat ecosystem services that reflected the ambient condition of the habitat such as water quality, erosion control and pollination ability that resulted in an end product; and row four, a cultural ecosystem service related to wetland biodiversity and identity values. Table 2a explains the 206 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 Table 2a Choice grid conditions in relation to state/price increments presented to core area respondents. Artisanal fisherfolk Natural(10,500,5000 VND annual tax) Landings (kg/months) 450 Due to little human activity and strict environmental regulations the habitat is able to deliver higher fish biomass increasing landings. 6 Aquatic Biodiversity (no. of species) Protected habitat richer in biodiversity with higher trophic organisms of greater market value (e.g. Pisodonophis boro and Lates calcarifer). Forest material delivery/access (honey, medicines, game/fruits) (min. by boat) Firefly sighting (monthly duration) 15 Time to encounter goods is shorter due to low development and conservation efforts. (Conditions are more efficient in maintaining services such as pollination, primary production, water purification). 6 Sightings are more frequent and lasts, hypothetically, two more months due to the improved environmental conditions and better conservation efforts. Current (7,000,000 VND annual tax) Alternative (8,750,000 VND annual tax) 300 150 Established after key informant interviews. Due to intensive aquaculture activity there is a lower biomass, decreasing landings. 4 Aquatic biodiversity is not exceptionally rich with particularly high market value organisms (organisms such as Mugil cephalusand certain gastropods). 30 A hypothetical median time between the natural and alternative state. 2 Intensive aquaculture activity has resulted in environmental degradation and cannot support higher trophic organisms. Only certain gastropods and molluscs of low market value are present. 45 Due to intensive aquaculture activity, regulating processes are impeded and the habitats ability to maintain and provide goods are degraded, resulting in longer encounter times. 4 2 Established after key informant interviews. Due to loss of mangroves as a result of intensive aquaculture, the surrounding environment cannot support large firefly populations and sightings are reduced. Mollusc farmers Natural (10,500,5000 VND annual tax) Current (7,000,000 VND annual tax) Alternative (8,750,000 VND annual tax) Harvest Frequency (months) 6 Harvest frequency is higher due to improved ambient conditions (water quality, suitable nutrient levels) and the crop is hypothetically larger in size. 4 Established after key informant interviews where most farmers harvest four times per crop cycle. Aquatic biodiversity (no. of species) 6 Protected habitat richer in biodiversity with higher trophic organisms of greater market value (e.g. Pisodonophis boro and Lates calcarifer). 4 Aquatic biodiversity is not exceptionally rich with particularly high market value organisms (organisms such as Mugil cephalusand certain gastropods). Forest material delivery/access (honey, medicines, game/fruits) (min. by boat) 15 Time to encounter goods is shorter due to low development and conservation efforts. (Conditions are more efficient in maintaining services such as pollination, primary production, water purification). 6 Sightings are more frequent and lasts, hypothetically, two more months due to the improved environmental conditions and better conservation efforts. 30 A hypothetical median time between the natural and alternative state. 2 Harvest frequency is lower due to degraded ambient conditions (water quality, suitable nutrient levels) as a result of intensive aquaculture. Crop is also hypothetically smaller in size. 2 Intensive aquaculture activity has resulted in environmental degradation and cannot support higher trophic organisms. Only certain gastropods and molluscs of low market value are present. 45 Due to intensive aquaculture activity, regulating processes are impeded and the habitats ability to maintain and provide goods are degraded, resulting in longer encounter times. Firefly sighting (monthly duration) 4 2 Established after key informant interviews. Due to loss of mangroves as a result of intensive aquaculture, the surrounding environment cannot support large firefly populations and sightings are reduced. Park rangers Natural (10,500,5000 VND annual tax) Aquatic biodiversity (no. of species) Ecotourism (visitor types) Costal Erosion (level of management urgency) Firefly sighting (monthly duration) Current (7,000,000 VND annual tax) Alternative (8,750,000 VND annual tax) 2 Intensive aquaculture activity has resulted in environmental degradation and cannot support higher trophic organisms. Only certain gastropods and molluscs of low market value are present. Local and foreign Local Minimal Intensive aquaculture has impacted the area Due to improved conservation and awareness efforts, Hypothetical scenario where only local the biosphere attracts significant local and foreign nearby tourists (namely Ho Chi Minh City) whereby almost no tourists visit the Can Gio tourists. More activities result in higher revenues. visit Can Gio mostly over the weekend. biosphere for recreational purposes. Low Moderate High Conservation efforts has resulted in increased Erosion is not an issue of priority due to Due to intensive aquaculture and mangrove mangrove cover and density, subsequently making insignificant impact. Business resumes as conversion erosion and land loss is displacing coastal erosion a ‘no issue’ for park authorities. usual for park authorities. local population and is an issue of importance for park authorities. 6 4 2 Sightings are more frequent and lasts, hypothetically, Established after key informant interviews. Due to loss of mangroves as a result of intensive two more months due to the improved environmental aquaculture, the surrounding environment conditions and better conservation efforts. cannot support large firefly populations and sightings are reduced. 6 Protected habitat richer in biodiversity with higher trophic organisms of greater market value (e.g. Pisodonophis boro and Lates calcarifer). 4 Aquatic biodiversity is not exceptionally rich with particularly high market value organisms (organisms such as Mugil cephalusand certain gastropods). S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 207 Table 2b Choice grid conditions in relation to state/price increments presented to buffer zone respondents. Shrimp farmers Natural (25% increase in production costs) Stocking Density (post larvae/m2) Aquatic biodiversity (no. of species) Disease outbreak (stock mortality and product application) Firefly sighting (monthly duration) 2 60/m Farmers stock at lower densities with most larvae growing to greater lengths, resulting in potential higher profits but with the consequence of lower production output per farm cycle. 6 Protected habitat richer in biodiversity with higher trophic organisms of greater market value (e.g. Pisodonophis boro and Lates calcarifer). 20% When an outbreak occurs the surrounding water quality is appropriate to mitigate pathogen outbreak. Farmers mitigate crop mortality by recovering the infected crop, exchanging water from adjacent canals and naturally treating the infected stock, loosing 20% of the infected crop with minimal product application. 6 Sightings are more frequent and lasts, hypothetically, two more months due to the improved environmental conditions and better conservation efforts. Current (No change in production costs) 2 Alternative (50% increase in production costs) 100/m Established after key informant interviews. 140/m2 Farmers stock at higher densities where larvae will be smaller in size, but due to higher stocking densities, profits may be higher greater due to high production. 4 Aquatic biodiversity is not exceptionally rich with particularly high market value organisms (organisms such as Mugil cephalusand certain gastropods). 2 Intensive aquaculture activity has resulted in environmental degradation and cannot support higher trophic organisms. Only certain gastropods and molluscs of low market value are present. 90% When an outbreak occurs the ambient water quality is poor due to the intensive shrimp aquaculture activity and most of the infected stock will not survive despite the high use of treatment products. 50% When an outbreak occurs during a crop cycle, the farmer will lose 50% of the infected crop after treating it with medium product applications. 4 Established after key informant interviews. three choice grids and its respective conditions presented to core area respondents and Table 2b the choice grid presented to buffer zone participants. 2.4. Choice selection analysis Relationships between respondent choice selection outcomes were tested with a Fisher’s exact test for each study area dataset. The chi-square test was not utilized as certain proxy selections had values below five (Upton, 1992). The Fisher’s exact tests determined the strength of the relationship between the different ecosystem service selections under their categorical variable; in this case the price increment/state and its degree of significance. In addition to determining the strength of a choice’s influence on a specific dataset, Pearson’s residuals were calculated. Pearson’s residuals indicate the difference between the observed and fitted values, over the estimated value of the standard deviation of the observed values. When residuals are greater than two in absolute value, the ecosystem service selection has had a significant influence on the final p-value and the overall distribution of the dataset. These relationships are visually represented by mosaic charts (Hartigan and Kleiner, 1981) along with Pearson’s residual distribution proportions showing the degree of significance of each ecosystem service selection. Along with rectangle sizes, color shades are utilized to represent Pearson’s residual derivatives and the ecosystem service’ level of significance on choice distributions. Grey indicates no significance. White specifies cross-classifications that occurred more often than the expected value, positively influencing distributions; black indicates cross-classification that occurred less than the expected value, negatively influencing data distribution. All the mentioned tests were conducted with the R 2.14.2 statistical package software. 3. Results 3.1. Ecosystem service values Total ecosystem service values for natural, current and alternative states at pre- and post-intervention are visually represented as 2 Due to loss of mangroves as a result of intensive aquaculture, the surrounding environment cannot support large firefly populations and sightings are reduced. percentage proportions in radar plots (Fig. 4). Specific core area selections revealed greater values placed on ecosystem services delivered atthe natural aswell asthe currentstate.Bufferzone groups made similar selections with strong initial preferences for ecosystem services delivered at the natural state and the current state. Significantly, price increments at both a 25% and a 50% increase became insignificant factors of consideration in both study groups during pre-intervention choice selection and even more so, postintervention. Core area preference distributions displayed strong valuesforecosystem services maintained at a natural state underpre(45%) and post- (56%) intervention deliberations. Preferences for ecosystem services maintained under the presented current state conditions revealed veryslight changesinpre-intervention (36%) and post-intervention (35%) selection. Few core area selections were made for ecosystem services maintained under the alternative state with pre-interventionpreferences low (18%) and a decrease observed for post-intervention selections (9%). However, a small proportion (9%) of the alternative state selections still endured post-intervention, indicating that particular core area respondents desired some form of aquaculture development benefits despite being aware of the risk of natural ecosystem service loss/replacement. Specific buffer zone focus groups selections revealed strong preferences, pre-intervention (66%) for ecosystem services delivered at the natural state with an increase (74%) in preferences, post-intervention. Values were placed for ecosystems services to be delivered at the current state (22%), pre-intervention with a slight decrease (21%), post-intervention. Very little value was expressed for ecosystem services delivered at the alternative state (12%) pre-intervention, with an observed decrease (6%) for postintervention selections. Significant values were placed on the buffer zone specific regulating ecosystem service of water quality maintenance where no selections were made under the alternative state, pre- and post-intervention, despite being aware of the potential costs and benefits. The indirect provisioning service of aquatic biodiversity was also highly valued by Can Gio shrimp farmers (73%) for its delivery under the natural state, preintervention. Direct provisioning selections on stocking density revealed that Can Gio shrimp farmers preferred stocking densities 208 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 Fig. 4. Radar plot of total preference percentage distribution, pre- (full line) and post-intervention (dashed line) for core area and buffer zone groups. at lower numbers, pre-intervention, where a difference of 12% is exhibited from the alternative or current state to the natural state, post-intervention. In both study groups, cultural service selections were more varied between the three presented states indicating the anecdotal manner attached to this proxy of firefly sightings. Post-intervention values revealed that 55% of the respondent shrimp farmers preferred cultural ecosystem services at the natural state and 31% preferred it at the current state. Interestingly, a small proportion of shrimp farmers still declared slight preferences for ecosystem services at the alternative state, postintervention, particularly for the direct provisioning service (stocking density) and cultural service (firefly sighting) proxies. 3.2. Mosaic plot Mosaic plots display choice selection influences on the final selected outcome distributions (Fig. 5). Core area pre-intervention choice selections are statistically different (p < 0.05), whereas post-intervention choice selections were not (p > 0.05). Core area post-intervention selection tests reveal little changes in Pearson’s residuals that significantly impacted choice distributions and although core area post-intervention selections shifted more towards the natural state, mosaic distributions reveal little proportional statistical changes occurring postintervention. For core area respondents the indirect provisioning service (aquatic biodiversity) positively influences choice distributions under the natural state (x2 = 2.06) and under the alternative state negatively influences final distributions (x2 = 2.06). Other core area residuals did not have significant consequences on the final distribution. Interestingly, cultural service residuals were more equal in distributions throughout the three states, where the alternative residual positively influenced choice distributions, pre-intervention (x2 = 1.88). Based on the pre-intervention residuals, combined with group debate outcomes, core area aversion for aquatic biodiversity to be maintained at the alternative state and respondent high preferences under a natural state significantly influenced final core area ecosystem service values, potentially overriding respondent desires of gaining potential aquaculture development benefits. Buffer zone pre- and post-intervention choice selections were significantly related (p < 0.05) where the regulating ecosystem service (water quality maintenance) significantly influenced final choice distributions positively at the natural state (x2 = 3.57) and negatively at the current state (x2 = 3.36) and the alternative state (x2 = 3.77). Residuals for the indirect provisioning ecosystem service (aquatic biodiversity) observed no significant statistical influences as heavy preference was initially declared for it at a natural state with little changes, post-intervention. A significant observation is the higher preference for the indirect provisioning service of aquatic biodiversity residual under the alternative state becoming a negative value (x2 = 2.15) after the intervention exercise. The direct provisioning service presented (stocking density) observed positive values under the current state (x2 = 2.68) revealing that Can Gio shrimp farmers are content to farm at current densities despite being presented with the opportunities to change stocking density and culture methods. Cultural ecosystem service value residuals (x2 = 2.89) under the natural state negatively influenced final distributions, whereas alternative residuals (x2 = 4.70) positively shaped final distributions. Cultural service distributions observed changes as shifts in values occurred from a current and alternative state to its improved delivery at the natural state. 3.3. Round up discussion outcomes Issues debated in the round up discussions were compiled into frequency percentages to obtain an overview on topical ecosystem service values discussed in both groups (Fig. 6). All core area groups discussed how unregulated shrimp aquaculture could negatively impact wetland habitats with some respondents also cautious of the outcomes of excessively conserving natural areas and its implications on their livelihoods (70%). The regulating ecosystem service of storm/erosion mitigation was a constant theme for most core area focus groups (70%) with less common issues such as how current natural resources do not provide enough for subsistence (20%) and that management efforts require stronger managerial implementation (20%) were debated in core area focus groups. Buffer zone focus groups discussed issues related to how shrimp farming activity can alter ecosystem service delivery and also revealed that historical experiences with shrimp farming, in particular, negative experiences with shrimp disease outbreaks have influenced shrimp farmers to value ecosystem services delivered at the natural state more. Significantly, most shrimp farmers highlighted that the value of living in Can Gio is characterized by the clean and natural area (81%); that rapid and heavy urbanization is something the community would like to avoid. Most shrimp farmers agreed on the negative impacts of S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 209 Fig. 5. Pre- and post-intervention core area and buffer zone mosaic plots of respondent ecosystem service choice selections. The Y variable is indicative of the ecosystem service conditions in relation to the X variable of state/price increment parameters. Rectangle proportions represent the choices selected. Rectangle heights are fixed. Lengths were subject to unique respondent choices. A significant observation of the mosaic plot is that buffer zone regulating/habitat ecosystem service values are conditionally dependent on the natural state price increment/conditions with a higher probability of preferences to be selected under the presented natural state. Fig. 6. Discussion frequencies of key topics during round up discussion in core area (n = 10) and buffer zone groups (n = 32). 210 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 Fig. 7. (a) Cross cutting themes discussed by both study groups and its degree of importance (1 = highest; 5 = lowest). (b) Total frequency of discussion topic distribution percentages based on comprehension ratings and type of study group. intensive shrimp aquaculture but also stressed how it has helped to improve the local economy (70%). Farmer groups also declared that aquaculture is necessary for an area like Can Gio to develop and improve community livelihoods through access to modern day services and facilities (41%). Common topics debated by both study groups revealed and compared what were the most pressing collective issues (1 = highest order; 5 = lowest order) (Fig. 7a). Core area respondents were most concerned about the potential negative impacts of intensive shrimp farming and how it can influence their livelihoods, whereas shrimp farmers did not express as much concern. Interestingly, both groups highlighted the importance of storm/erosion protection and the presence of mangrove forests in delivering this service, often referring to a series of heavy storms in 2007, describing how mangrove forests helped mitigate damage in areas with higher mangrove cover and how other areas, with less mangrove cover suffered greater damage. Based on comprehension ratings, the five cross cutting issues were mentioned and discussed more often for both study groups when comprehension ratings had good or average quality (Fig. 7b). When comprehension ratings were of good or average quality, discussion topic frequency decreased and debate quality on ecosystem services in relation to shrimp aquaculture and wetland conservation were not deliberated in greater detail. 4. Discussion 4.1. Ecosystem service value determinants and its methodological consequences Natural wetland ecosystem services were highly valued and although both groups had different sets of initial values attached to wetland protection and shrimp aquaculture outcomes, intrinsic ecosystem services were declared irreplaceable. These included S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 services such as aquatic biodiversity wildlife for communal subsistence, storm/erosion protection and air/water quality control. Study results revealed that ecosystem service values stem from a user specific context with a final ecosystem service product. These values often undergo personal extrapolation based on several factors; for example, buffer zone respondents valued water quality and aquatic biodiversity higher than core area respondents. Study outcomes further suggest that ecosystem service values attached to ecosystem services are pre-determined by (1) the presence of natural landscapes, (2) the user specific historical attachments to ecosystem service end products, and (3) current access and interaction modes with ecosystem service functions. The three outcomes that highlight fixed values are: core and buffer zone high initial preferences for the aquatic biodiversity indirect provisioning service under the natural state; shrimp farmer strong initial preferences to the water quality regulating ecosystem service under the natural state, and park ranger selections of ecotourism and storm/erosion protection services delivered at the natural state. This predisposition to prefer certain ecosystem services influence how users value natural habitats and is the major difficulty in accurate and holistic contingent valuation methods (Pearce et al., 2006; Schägner et al., 2013). Historical experiences also played an important role in ecosystem service value formulation. In the study shrimp farmer respondents discussed the negative impacts of past necrotizing hepatopancreatitis outbreaks on production and these negative historical experiences, combined with the high-risk modis operandi associated with intensive shrimp farming may well have fixed buffer zone regulating ecosystem service values, explaining their explicit preferences for ecosystem services delivered under the presented natural state. Norton et al. (1998) and Costanza (2000) discuss how ecosystem service values become fixed and pre-formed at the local level on an efficiency value basis. The study results also suggest that user interaction modes with ecosystem service functions and eventually ecosystem service end products are determined on an efficiency basis. This factor transpired during group deliberations and is also documented in other ecosystem service valuation studies (Spash and Hanley, 1995; Nunes and van den Bergh, 2001; Hein et al., 2006; Kenter et al., 2011). By attempting to incorporate contrasting ecological and economic drivers with distinct habitat outcomes, the method has consequently drawn out the complexities that occur when attaching monetary values to ecosystem services in a multi functional landscape (Do and Bennett, 2005; Kenter et al., 2011. Therefore, it is important to disaggregate user specific connections to ecosystem service functions and the end product/goods; where the users will begin to attach values. This distinction would be an improvement in holistically assessing habitat value for indirect and direct use and non-use services (Barkmann et al., 2008). Further, oversighting between ecosystem service functions and products can be made more complex when respondents perceived threats to an intrinsic ecosystem services, i.e., intensive shrimp farming degrading water quality. Once threats are triggered it becomes increasingly difficult to associate ecosystem services to a practical monetary value whereby certain regulating and supporting products become un-substitutable as it constitutes the users perception of a pristine habitat (Hall et al., 1992; Carson and Mitchell, 1993). This factored with the importance of spatial and historic interactions raises further issues on contingent valuation methods, especially for regulating and supporting ecosystem services who’s functions tend to be more readily associated intrinsically to end products such as water/air quality maintenance (Hein et al., 2006; Simões d et al., 2008). When natural habitats and human social contexts and determinants are variable, significant influences on valuation 211 accuracies may occur (Nelson and Daily, 2010; Schägner et al., 2013; Ellis and Bell, 2013). In monetary valuation studies, natural value is defined as the human worth attached to accessing and utilising natural resources (Hargrove, 1992; Turner, 2000) and essentially, the stronger the values declared for an ecosystem service, it becomes less realistic to associate the service with a monetary price, especially when the ecosystem service is communally deliberated. Costanza (2000), Moons et al. (2008), and Schägner et al. (2013) support this contextual significance whereby efficiency values, rarity of the presented ecosystem service, or if certain ecosystem services have been lost or replaced will determine how users perceive its worth. Industrialised shrimp farms were not observed in the study location as all respondent farms were family owned. However, shrimp farming in southern parts of Vietnam (Ca Mau, Bac Lieu, and Soc Trang), large private companies operate intensive shrimp farming clusters (World Atlas of Mangroves, 2010). The extent of industrialised companies operating may have uniquely moulded ecosystem service values for different communities. Based on factors such as production efficiency, environmental awareness, social corporate responsibilities and general management coupled with socio-economic and biophysical components, the involvement of industrialised companies would alter community ecosystem service values and essentially attitudes on conservation and development. Ecosystem service provides an approach to holistically manage human landscapes but is limited to large geographic areas and long time scales. It is therefore equally important to select the most appropriate methods to identify and assess a communities’ degree of initial connection and relationship with an industry. Both respondent group outcomes support individual utility maximization (Norton et al., 1998; Costanza, 2000). This initial utility value stems from the personal desire to gain the most out of a natural habitat that is translatable into a monetary number, with minimal impact to local ecosystem service bundles. Ekins and Max-Neef (1992) and Farber et al. (2002) suggest that some ecosystem service values and the human interactions are ‘purchasable’ but also declare that certain human needs may not be satisfied through capital, as individuals simply may not consider its loss/replacement as worthy beyond a capital value. Capital was a minor influencing factor and despite being constantly reminded to consider price increments, it seldom affected respondent choices and became less significant, post-intervention, in shaping respondent ecosystem service values. This lack of observed consideration to price increments in both study groups support how intrinsic values will supersede the monetary connections to natural ecosystem services where marginal price increments seemingly acted to stimulate the respondents to readily declare a strong willingness to accept payments for storm protection and water quality ecosystem services (Chee, 2004; McCauley, 2006). However, deliberative approaches go some way to override inaccuracy issues associated with monetary ecosystem service valuations and its typology (Turner et al., 2003; Kumar and Kumar, 2008) but are more subject to double counting inaccuracies based on its lexicographic manner (Fu et al., 2011). 4.2. Deliberation impacts on community ecosystem service values and learning outcomes Respondents debated on ecosystem service functions and many participants were able to declare final products such as the provisioning of shrimp larvae, clean water to culture bivalves, and the sedimentation process as key indicators (Sasikumar and Krishnakumar, 2011) but were unable to expand on how ecosystem services help deliver products and how they are potentially impacted by shrimp aquaculture and wetland conservation. 212 S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 Respondent reception to intensive farming and conservation were transformed after the intervention exercise, highlighting the importance of awareness, education and deliberation in influencing ecosystem service values at the local level, particularly in groups (Wilson and Howarth, 2002). After the intervention exercise, participants better understood how aquaculture and wetland conservation outcomes influence ecosystem service functions with most respondents agreeing that aquaculture development was essentially a means to install a market mechanism, potentially leading to short-term infrastructure and employment improvements. With deliberation influencing final respondent choice outcomes, it is important that valuation methods are conducted through communal deliberative approaches of ecosystem service functions for its improved mainstreaming into wetland management. This will help achieve more equitable and less environmentally detrimental industries by addressing local values from the ground up (Heal et al., 2001; Armsworth et al., 2007). Intervention and deliberation outcomes for both study groups drew out a stronger sense of altruism that prevailed over economic gains to preserve communal ecosystem services functions, specifically supporting and regulating ecosystem services delivered under the natural state (Sayer and Campbell 2004; Cowling et al., 2008). Locally expressed as , this connection with natural functions and human communities eventually became more valuable than short-term economic gains when presented with the potential loss/ replacement of ecosystem services which development ventures impact (Sagoff,1998; Wilson and Howarth, 2002; Kumar and Kumar, 2008). Farber et al. (2002) emphasized that some ecosystem service functions are so isolated from human economic relation that they become irrelevant and only the intrinsic connection remain, holding only indirect values in a market economy. Ecosystem services are delivered in bundles and their indirect values should not be overlooked as they often resurface when deliberated, sometimes with unlimited worth, especially when they are perceived to be under threat or lost at the communal local level. These vast range of unique learning outcomes discussed by the respondent groups underlines the importance of a participatory approach when mapping ecosystem service values (Reed, 2008). However, it also raises further methodological issues for its applications in economically valuing multifunctional areas to a practical estimate, especially with the existing linkage gaps between ecosystem services, biodiversity and land use (Chan et al., 2006), particularly in rural areas. Asides from Cardinale (2011); the direct roles of biodiversity in ecosystem service resilience, societal value and ecosystem service functionality is poorly understood and integrated. In addition, tipping points and social threshold perceptions of when landscapes shift into a less desired state as a consequence of human actions and its level of acceptance by local communities is another research dimension gap (Schägner et al., 2013). This study attempted to address this through the use of ecosystem services, price increments and deliberative scenario outcomes but have revealed that these elements are perplexed when different stakeholders, histories, and spatial scales are encompassed (Diamond and Hausman, 1994; Hanemann, 1994). Furthermore, the importance of habitat stewardship in influencing final choice selection should be quantified. Both user groups claimed that other more detrimental players such as upstream industries, shipping, and destructive fishing techniques are the significant stressors impacting Can Gio’s natural habitat. Many buffer zone respondents strongly believed that upstream industries, not shrimp farming are the main drivers influencing biodiversity and ecosystem service capacities to deliver goods/ services under a natural state. This stems from discussion outcomes with shrimp farmers whereby many collect mud crab (Scylla serrata) and many declared that catch biomass, crab sizes, and catch frequency have decreased over the past five years. Another example stated by park rangers discussed how heavy shipping routes, not aquaculture, is greatly accelerating land erosion in Can Gio. Buffer zone respondents frequently discussed land ownership issues and declared that much of the actual land/ farm owners are not Can Gio locals and tend to be more business orientated, paying less attention to the consequences of their actions/inactions. These examples highlight the how stewardship can also influence and shape the basis of the users connection to ecosystem service worth (Farber et al., 2002). Conclusion Traditional valuation methods attempt to place monetary figures on use values, possibly overlooking non-use values that may be of equal or even greater worth to local communities (Hampicke, 1999; Kenter et al., 2011). The study pinpoints the methodological issues associated with ecosystem service deliberations and how intrinsic values take shape through group discussion. Both core and buffer zone groups strongly valued ecosystem services delivered at the natural state, where certain stakeholders declared high values for regulating, supporting and provisioning ecosystem services such as aquatic biodiversity, ambient conditions to mitigate shrimp disease outbreaks, erosion/ storm protection and deliverer of clean air/water; cultural values were more anecdotal. Focus group discussions revealed that price increments are issues of insignificance prior the intervention exercise and even more so, post-intervention; that certain intrinsic ecosystem services become extremely difficult to associate with a monetary number influenced by a landscape's unique socioeconomic and biophysical historical dimensions which are often robust and pre-formed. The lexicographic manner of landscape valuation should not render our attempts to monetarily value habitats. Application of ecosystem services in economic valuation methods will result in a more rounded approach to habitat management in light of different social and economic drivers and clarify why respondents might sacrifice or protect biodiversity ecosystem services. The future need to better incorporate the role of ecosystem service functions and biodiversity in valuation methods is required to address the uncertainties in landscape valuation with communal deliberation and discussion an integral part of the process. Acknowledgments The authors would like to thank the Royal Thai Government and the Swedish International Development Cooperation Agency (Sida) for funding the study and fieldwork. Gratitude is also expressed to the Can Gio Park Management Authorities and the Can Gio’s People’s Committee for logistic and personnel assistance. References Armsworth, P.R., Chan, K.M.A., Dailey, G.C., Ehrlich, P.R., Kremen, C., Ricketts, T.H., Sanjaya, M.A., 2007. Ecosystem service science and the way forward for conservation. Conser. Biol. 21 (6), 1383–1384. Ashton, E.C., 2007. The Impact of shrimp farming on mangrove ecosystems, CAB reviews: perspectives in agriculture, veterinary science. Nutr. Nat. Res. 3 (003) . Bakker, K., 1999. The politics of hydropower: developing the Mekong. Polit. Geogr. 18, 209–232. Barkmann, J., Glenk, K., Keil, A., Leemhuis, C., Dietrich, N., Gerold, G., Marggraf, R., 2008. Confronting unfamiliarity with ecosystem functions: the case for and ecosystem service approach to environmental valuation with stated preference methods. Ecol. Econ. 65 (1), 48–62. Can Gio Annual Report, Annual shrimp production figures in Can Gio districts. Can Gio’s Peoples Committee Report Bin Kahn. 2011; [In Vietnamese]. Cardinale, B.J., 2011. Biodiversity improves water quality through niche partitioning. Nature 472, 86–89. S. McDonough et al. / Ecological Indicators 46 (2014) 201–213 Carson,R.T.,Mitchell,R.C.,1993. The Value ofcleanwater: the public’s willingnessto pay for boatable, fishable and swimmable water. Water Resour. Res. 29 (7),1454–2445. Chan, K.M.A., Shaw, M.R., Cameron, D.R., Underwood, E.C., Daily, G.C., 2006. Conservation planning for ecosystem services. PLoS Biol. 4 (11), 379. Chee, Y.E., 2004. An ecological perspective on the valuation of ecosystem services. Biol. Conserv. 120, 549–565. Costanza, R., 2000. Social goals and the valuation of ecosystem services. Ecosystems 3, 4–10. Cowling, R.M., Egoh, B., Knight, A.T., O'Farrell, P.J., Reyers, B., Rouget, M., Roux, D. J., Welz, A., Wilhelm-Rechman, A., 2008. An operational model for mainstreaming ecosystem services for implementation. Proc. Nat. Acad. Sci. 105 (28), 9483–9488. Diamond, P.A., Hausman, J.A., 1994. Contingent valuation: is some number better than no number? J. Econ. Perspect. 8 (No. 4), 45–64. Do, T.N., Bennett, J.,2005.Economic valuation ofwetlandsinVietnam’s MekongDelta:a case study of direct use values in Ca Mau province. Occasional Paper, Environment Management and Development Program. APSEG ANU, Canberra, Australia. EJF, 2003. Risky Business: Vietnamese Shrimp Aquaculture – Impacts and Improvements. Environmental Justice Foundation, London, UK. Ekins, P., Max-Neef, M., 1992. Real Life Economics. Routledge, London. Ellis, W.L., Bell, S.S., 2013. Intertidal fish communities may make poor indicators of environmental quality: lessons from a study of mangrove habitat modification. Ecol. Indic. 24, 421–430. FAO, 2011. FishStatJ (Version 2.0.0) Vietnam giant tiger prawn and white leg shrimp aquaculture production quantitates 1950-2011. FAO Fisheries department, fishery information, data and statistic unit. FAO fisheries and aquaculture department, Rome. Farber, S.C., Costanza, R., Wilson, M.A., 2002. Special Issue: The dynamics and value of ecosystem services: integrating economic and ecological perspectives. Economic and ecological concepts for valuating ecosystem services. Ecol. Econ. 41, 375–392. Frankic, A., Hershner, C., 2003. Sustainable aquaculture: developing the promise of aquaculture. Aquac. Int. 11, 517–530. Fu, B.-J., Su, C.-H., Wei, Y.-P., Lü, Y.-H., 2011. Double counting in ecosystem services valuation: causes and countermeasures. Ecol. Res. 26 (1), 1–14. Gunawardena, M., Rowan, J.S., 2005. Economic valuation of a mangrove ecosystem threatened by shrimp aquaculture in Sri Lanka. Environ. Manage. 36 (4), 535–550. Hall, J.V., Winer, A.M., Kleinman, M.T., Lurmann, F.W., Brajer, V., Colome, S.D., 1992. Valuing the health benefitsof clean air. Science 255 (5046), 812–817. Hampicke, U., 1999. The limits to economic valuation of biodiversity. Int. J. Soc. Econ. 26 (1/2/3), 158–173. Hanemann, M.W., 1994. Valuing the environment through contingent valuation. J. Econ. Perspect. 8 (4), 19–43. Hanley, N., Spash, C.L., 2001. Cost–benefit analysis and environmental policymaking. Environ. Plan. C: Gov. Pol. 19 (1), 103–118. Hargrove, C., 1992. Weak anthropocentric intrinsic value. Monist. 75, 183–207. Hartigan, J.A., Kleiner, B., 1981. Mosaics for contingency tables. Computer Science and Statistics: Proceedings of the 13th Symposium on the Interface 268–273. Heal, G., Dailey, G.C., Ehrlich, P.R., Salzman, J., 2001. Protecting natural capital through ecosystem services. Stanford Environ. Law J. 20, 333. Hein, L., Koppen v, K., de Groot, R., van Ierland, E.C., 2006. Spatial scales, stakeholders and the valuation of ecosystem services. Ecol. Econ. 57, 209–228. Hori, H., 1993. Development of the Mekong river basin, its problems and future prospects. Water Int. 18 (2), 110–115. Kenter, J.O., Hyde, T., Christie, M., Fazey, I., 2011. The importance of deliberation in valuing ecosystem services in developing countries – Evidence from the Solomon Islands. Glob. Environ. Change 21, 505–521. Kumar, M., Kumar, P., 2008. Valuation of the ecosystem services; a psycho-cultural perspective. Ecol. Econ. 64, 808–819. McCauley, D.J., 2006. Selling out on nature. Nature 443, 27–28. MEA, 2005. The United Nations Millennium Ecosystem Assessment. Ecosystems and Human Well-being: Current State and Trends. Coastal systems. Island Press, Washington, D.C. USA. 213 Moons, E., Saveyna, B., Proosta, S., Hermy, M., 2008. Optimal location of new forests in a suburban region. J. Forest Econ. 14, 5–27. Nelson, E.J., Daily, G.C., 2010. Modelling ecosystem services in terrestrial systems. F 1000 Biol. Rep. . Norton, B., Costanza, R., Bishop, R., 1998. The evolution of preferences: why “sovereign” preferences may not lead to sustainable policies and what to do about it. Ecol. Econ. 24, 193–211. Nunes, P.A.L.D., van den Bergh, J.C.J.M., 2001. Economic valuation of biodiversity: sense or nonsense? Ecol. Econ. 39, 203–222. Páez-Osuna, F., 2001. The Environmental impact of shrimp aquaculture: causes, effects, and mitigating alternatives. Environ. Manage. 28 (1), 131–140. Pearce, D., Atkinson, G., Mourato, S., 2006. Cost Benefit Analysis and the Environment: Recent Developments. Publishing OECD, Paris, France. Primavera, J.H., 1997. Fish predation on mangrove-associated penaeids, the role of structures and substrate. J. Exp. Mar. Biol. Ecol. 215, 205–216. Reed, M.S., 2008. Stakeholder participation for environmental management: a literature review. Biol. Conserv. 141 (10), 2417–2431. Rönnbäck, P., 1999. The ecological basis for economic value of seafood production supported by mangrove ecosystems. Ecol. Econ. 29, 235–252. Rosenberry, B., 1998. World Shrimp Farming. Shrimp News International, San Diego. Sagoff, M., 1998. Aggregation and deliberation in valuing environmental public goods: a look beyond contingent pricing. Ecol. Econ. 24, 213–230. Sasikumar, G., Krishnakumar, P.K., 2011. Aquaculture planning for suspended bivalve farming systems: the integration of physiological response of green mussel with environmental variability in site selection. Ecol. Indic. 11, 734–740. Sayer, J.A., Campbell, B.M., 2004. The Science of Sustainable Development. Local Livelihoods and the Global Environment. Cambridge University Press, Cambridge, U. K. Schägner, J.P., Brander, L., Maes, J., Hartje, V., 2013. Mapping ecosystem services’ values: current practice and future prospects. Ecosyst. Serv. 4, 33–46. Simões d, F.S., Moreira, A.B., Bisinoti, M.C., Gimenez, S.M.N., Yabe, M.J.S., 2008. Water quality index as a simple indicator of aquaculture effects on aquatic bodies. Ecol. Indic. 8, 476–484. Spash, C.L., Hanley, N., 1995. Preferences, information and biodiversity preservation. Ecol. Econ. 12, 191–208. TEEB, 2010. The Economics of Ecosystems and Biodiversity: The Ecological and Economic Foundations. Earthscan, London. Thuy, T.D., 2006. Willingness to pay for conservation of Vietnamese rhino. Paper presented at the EEPSEA biannual workshop November . Turner, R.K., 2000. The place of economic values environmental valuation. In: Batman, I.J., Willis, K.G. (Eds.), Valuing Environmental Preferences. Oxford University Press, pp. 19–41. Turner, R.K., Paavola, J., Cooper, P., Farber, S., Jessamy, V., Georgiou, S., 2003. Valuing nature: lessons learned and future research directions. Ecol. Econ. 46 (3), 493– 510. UNESCO, 2000. Biosphere Reserve Information: Vietnam – Can Gio Mangrove. UNESCO MAB Biosphere Reserves Directory. Upton, J.G., 1992. Fisher’s exact test. J. R. Statist. Soc. A 155 (3), 395–402. Wilder, M.N., Phuong, N.T., 2002. The status of aquaculture in the Mekong delta region of Vietnam: sustainable production and combined farming systems. , Fisheries ScienceProceedings of International Commemorative Symposium 70th Anniversary of the Japanese Society of Fisheries Science 70th Anniversary of the Japanese Society of Fisheries Science, 68 Suppl. I. . Wilson, M.A., Howarth, R.B., 2002. SPECIAL ISSUE: The dynamics and value of ecosystem services: integrating economic and ecological perspectives. Discourse-based valuation of ecosystem services: establishing fair outcomes through group deliberation. Ecol. Econ. 41, 431–443. World Atlas of Mangroves, 2010. Chapter: South East Asia. Earthscan, Vietnam, pp. 129–131. World Bank, 1999. Project Appraisal Document on a Proposed Credit to the Socialist Republic of Vietnam for a Coastal Wetlands Protection and Development Project. Report No: 19825 VN. Rural Development and Natural Resources Sector Unit, East Asia and Pacific Region.