an ecosystem-based approach and management framework for

AQUACULTURE ENVIRONMENT INTERACTIONSAquacult Environ Interact

Vol. 2: 193–213, 2012doi: 10.3354/aei00040

Published online May 8

© Inter-Research 2012 · www.int-res.com*Email: [email protected]

An ecosystem-based approach and managementframework for the integrated evaluation of bivalve

aquaculture impacts

Peter J. Cranford1,*, Pauline Kamermans2, Gesche Krause3, Joseph Mazurié4, Bela H. Buck5, Per Dolmer6, David Fraser7, Kris Van Nieuwenhove8, Francis X.

O’Beirn9, Adoración Sanchez-Mata10, Gudrun G. Thorarinsdóttir11, Øivind Strand12

1Dept. of Fisheries and Oceans, Bedford Institute of Oceanography, PO Box 1006, Dartmouth, Nova Scotia B2Y 4A2, Canada2Wageningen Institute for Marine Resources and Ecosystem Studies, PO Box 77, 4400 AB Yerseke, The Netherlands

3Center for Tropical Marine Ecology, Fahrenheitstr. 6, 28359 Bremen, Germany4IFREMER, LER/MPL, BP 86, 56470 La Trinité/mer, France

5Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, Germany6Technical University of Denmark, National Institute of Aquatic Resources, Kavalergårdden 6, 920 Charlottenlund, Denmark

7Fisheries Research Services, Marine Laboratory, PO Box 101, 375 Victoria Road, Aberdeen AB11 9DB, UK8Institute for Agricultural and Fisheries Research − Fisheries, Ankerstraat 1, 8400 Oostende, Belgium

9Marine Institute, Rinville, Oranmore, Galway, Ireland10Marine Resources Research Department, Marine Research Center, Xunta de Galicia, Aptdo. 13, Vilanova de Arousa 36620, Spain

11Marine Research Institute, Skulagata 4, PO Box 1390, 121 Reykjavik, Iceland12Institute of Marine Research, Nordnesgt. 50, Boks 1870 Nordnes, 5817 Bergen, Norway

ABSTRACT: An ecosystem-based approach to bivalve aquaculture management is a strategy forthe integration of aquaculture within the wider ecosystem, including human aspects, in such away that it promotes sustainable development, equity, and resilience of ecosystems. Given thelinkage between social and ecological systems, marine regulators require an ecosystem-baseddecision framework that structures and integrates the relationships between these systems andfacilitates communication of aquaculture–environment interactions and policy-related develop-ments and decisions. The Drivers-Pressures-State Change-Impact-Response (DPSIR) manage-ment framework incorporates the connectivity between human and ecological issues and wouldpermit available performance indicators to be identified and organized in a manner that facilitatesdifferent regulatory needs. Suitable performance indicators and modeling approaches, which areused to assess DPSIR framework components, are reviewed with a focus on the key environmentalissues associated with bivalve farming. Indicator selection criteria are provided to facilitate con-straining the number of indicators within the management framework. It is recommended that anecosystem-based approach for bivalve aquaculture be based on a tiered indicator monitoring sys-tem that is structured on the principle that increased environmental risk requires increased mon-itoring effort. More than 1 threshold for each indicator would permit implementation of predeter-mined impact prevention and mitigation measures prior to reaching an unacceptable ecologicalstate. We provide an example of a tiered monitoring program that would communicate knowledgeto decision-makers on ecosystem State Change and Impact components of the DPSIR framework.

KEY WORDS: Bivalve aquaculture management · Ecosystem-based approach · DPSIR framework ·Indicators · Thresholds · Benthic effects · Pelagic effects · Social-ecological systems

Resale or republication not permitted without written consent of the publisher

OPENPEN ACCESSCCESS

Aquacult Environ Interact 2: 193–213, 2012

INTRODUCTION

Aquaculture is the fastest growing food-producingsector in the world and is expected to continue togrow and help compensate for the anticipated globalshortage of supply from capture fisheries. Whilethere is a clear need for the continued worldwideexpansion of aquaculture to fill this gap, this devel-opment needs to be promoted and managed in aresponsible manner that minimizes negative envi-ronmental impacts. To ensure that human activitiesare carried out in a sustainable manner, numerousinternational maritime policies have been imple-mented, including the European Union ‘Water Frame -work’ and ‘Marine Strategy Directives,’ the Cana-dian ‘Oceans Act,’ the US ‘Ocean Action Plan,’ andthe Australian ‘National Strategy for Ecological Sus-tainable Development.’ These legislations and poli-cies mandate that decision making and marine man-agement should include as nested components: (1) aknowledge-based approach, (2) an ecosystem-basedapproach, and (3) an integrative management frame-work that includes economic, environmental, social,and equity considerations. The International Councilfor the Exploration of the Sea (ICES) Working Groupon Marine Shellfish Culture was tasked with provid-ing a recommended governance approach for theintegrated evaluation of the impacts of bivalve aqua-culture activities in the coastal zone. This paper sum-marizes our deliberations and recommendations.

The ecosystem-based management approach hasbeen defined as ‘a comprehensive integrated man-agement of human activities based on the best avail-able scientific knowledge about the ecosystem andits dynamics, in order to identify and take actions oninfluences that are critical to the health of ecosys-tems, thereby achieving sustainable uses of ecosys-tem goods and services and maintenance of ecosys-tem integrity’ (Rice et al. 2005, p. 4). The Conventionon Biological Diversity defined the ecosystem-basedapproach as a strategy for the integrated manage-ment of land, water, and living resources that pro-motes conservation and sustainable use in an equi-table way (www.cbd.int/ecosystem/; see also Soto etal. 2008). Such an approach must strive to balancediverse societal objectives (e.g. GESAMP 2001,Nobre et al. 2009) by taking into account the knowl-edge and uncertainties of biotic, abiotic, and humancomponents of ecosystems, in cluding their interac-tions, flows, and processes (eco system functions) andservices within ecologically and operationally mean-ingful boundaries. Three spatial scales/levels of anecosystem-based approach to aquaculture manage-

ment include the farm, the water body and its water-shed/ aquaculture zone, and the global, market-tradescale.

Bivalve aquaculture−environment interactions withina given water body have often been examined by scientists using the concept of ‘carrying capacity’(McKindsey et al. 2006, Gibbs 2007, Grant &Filgueira 2011). This approach has traditionallyfocused on predicting the maximum sustainableyield of the bivalve culture (i.e. production carryingcapacity) and therefore primarily reflects an eco-nomic management perspective. The more recentecocentric focus on the sustainability of an area foraquaculture development has led to some attempts toassess carrying capacity by considering potentialchanges in ecosystem structure and function andecological variability over different spatial and tem-poral scales. This ‘ecological carrying capacity’approach is defined as the level of culture that can besupported without leading to significant changes toecological processes, species, populations, or com-munities in the growing environment (Gibbs 2007).Although the modeling of ecological carrying capac-ity is still in its infancy, it is more directly linked to theobjectives of ecosystem-based management thanproduction carrying capacity modeling. However, tobe compatible with the philosophy of the ecosystem-based approach (i.e. a balance of ecological andhuman needs), an integration of all of the hierarchi-cal categories of carrying capacity (physical, produc-tion, eco logical, and social; McKindsey et al. 2006)is required.

Ecosystem-based aquaculture management re -quires many specific components and tools in addi-tion to models and tools for assessing the carryingcapacity of an area. These include methodologies,technologies, and policies for hazard identification,risk assessment and management, environmentalquality assessment, environmental monitoring pro-grams and associated sampling designs, impactassessment, impact mitigation, and decision supportand communication among stakeholders. In addition,an overall decision framework is required that willpermit the ecosystem-based management approachto be integrated with economic and social considera-tions and communicated to diverse stakeholders.Consequently, much of our science-based advice andrecommendations focused on the following topics:

(1) An operational management framework fordecision-makers that facilitates ecological sustain-ability by considering the capacity to incorporate anecosystem perspective, societal values, and the eco-nomic viability of industry.

194

Cranford et al.: Bivalve aquaculture management

(2) Effective performance-based approaches andindicators for characterizing ecosystem status changesand impacts from a highly diverse bivalve aquacul-ture industry.

(3) Identifying the potential consequences tocoastal marine ecosystems from aquaculture relatedto specific changes in environmental quality andimpacts and identifying related thresholds of poten-tial public concern.

While scientists have an important place in design-ing a framework for the management of bivalveaquaculture, it is not their responsibility to make finaldecisions about regulatory policies. A recurring bottle -neck to the establishment of an operational frame-work is the need to define ‘unacceptable’ impacts.While natural science has an important role in advis-ing managers and policy-makers on the ecologicalconsequences related to available managementoptions, the setting of impact limits needs to incorpo-rate societal values and needs and economic reali-ties. Although socio-economic issues were initiallyconsidered outside the scope of our activities, delib-erations on many components of a bivalve aquacul-ture management approach required discussion ofthe costs of implementing our advice to a highlydiverse industry and what impact level may consti-tute a ‘potential’ public concern. To help define whatlevel of impacts are acceptable, social sciences canhelp in clarifying the values and expectations of dif-ferent groups and contribute to the economic evalua-tion of environmental services known to be providedby bivalve aquaculture (e.g. provision of habitat,water clarification, shoreline stabilization; Cohen etal. 2011). International environmental conservationand protection legislations pertaining to the utiliza-tion of coastal areas generally reflect societal values,and an analysis of pertinent policy statements canprovide useful insights towards identifying regula-tory triggers/thresholds.

This paper is structured to present and rationalizeour recommendations by starting with brief over -views of potential ecological interactions with bivalveaquaculture and the attributes of available integra-tive management frameworks. The potential man-agement roles of ecological modeling and indicator-based approaches for describing ecosystem statusand aquaculture sustainability are then discussed.More specific aspects of a recommended bivalve aqua -culture management framework are then addressed,including the identification of performance indicatorsrelated to specific environmental, and to some extentsocio-economic, effects from bivalve culture opera-tions. This leads to a discussion on thresholds of eco-

logical and potential public concern, and some keyelements of an ecological monitoring program for thehighly diverse bivalve aquaculture industry.

ECOSYSTEM INTERACTIONS WITH BIVALVEAQUACULTURE

A fundamental understanding of the influence ofthis expanding industry on coastal ecosystems, aswell as interactions with other anthropogenic stres-sors, is the foundation for developing strategies forsustainable aquaculture and integrated coastal zonemanagement. The culture of bivalve mollusks andtheir associated rearing structures has the potentialto impact the environment in numerous positive andnegative ways (e.g. Dame 1996, Souchu et al. 2001,Christensen et al. 2003, Newell 2004, Cranford et al.2006, 2007, Dumbauld et al. 2009, Forrest et al. 2009,McKindsey et al. 2011, Shumway 2011). Environ-mental concerns regarding bivalve culture are relatedprimarily to how the culture interacts with, and po -tentially controls, fundamental ecosystem processes.Where negative effects have been reported, they aregenerally linked to the consumption of suspendedparticles and particularly the phytoplankton, effectson coastal nutrient dynamics from ammonia excre-tion and organic waste recycling, and effects result-ing from the translocation of suspended matter frompelagic to benthic compartments (Fig. 1). Positiveeffects on biodiversity can result from the intro -duction of additional biotic and abiotic structure tothe system, the increase or alteration of prey avail-ability (cultured and fouling species), the capacity ofbivalves to clarify water and extract excess phyto-plankton that can have harmful effects in eutrophicareas, and/or from enhanced seabed organic en -richment (Callier et al. 2008, D’Amours et al. 2008).Given the intensity of bivalve culture in some regionsand the complex nature of positive and negativeaquaculture–environment interactions, an ecosystem-based perspective is mandatory for assessing the netenvironmental impact (Cranford et al. 2006).

Research on bivalve culture impacts have primarilyfocused on the production and sedimentation oforganic biodeposits (feces and pseudofeces) that mayimpact benthic biogeochemistry and the structureand composition of benthic and pelagic communities.The effects of bivalve farms on benthic environmentsare relatively well known (see reviews by Cranfordet al. 2006, 2008, McKindsey et al. 2011, Shumway2011). Bivalves have an exceptional capacity to filterlarge volumes of water to extract phytoplankton and

195


other suspended particulate matter and can, undersome conditions, control suspended particle dynam-ics at the coastal ecosystem scale (e.g. Cloern 1982,Officer et al. 1982, Dame 1996, Grant et al. 2008).Bivalve filter-feeding naturally results in some localreduction (depletion) of their food supply (suspendedparticulate matter). If the bivalve culture is consum-ing the seston faster than it can be replaced by tidalflushing and phytoplankton growth, then the culturewill become food limited and farm yield will becomeless than maximal for that site. If the spatial scale ofphytoplankton depletion expands outward from thefarm(s) to include a significant fraction of the coastalinlet, then this effect on the base of the marine foodweb generates ecological costs to other componentsof the ecosystem that may result in significant eco-logical and socio-economic consequences. Potentialeffects on nutrient cycling, fluxes, and retention mayalso be expressed at the coastal ecosystem scale (e.g.Newell 2004, Nizzoli et al. 2006, Cranford 2007) suchthat the boundary of the aquaculture system to bemanaged must include both benthic and pelagiccomponents and extend beyond the footprint of thefarm.

The spatial extent and magnitude of ecologicalinteractions with bivalve culture are always site-specific, with vulnerability depending on factors con-trolling food consumption and waste production (e.g.intensity of culture and food supply) and waste dis-

persion. Waste dispersion rate is controlled by physi-cal factors (e.g. circulation, bathymetry, and coastalmorphology) and largely determines the capacity ofthe local environment to prevent excessive fooddepletion, altered nutrient dynamics, and benthicorganic enrichment.

INTEGRATIVE MANAGEMENT

Decisions underlying the design of an ecosystemapproach for bivalve culture should be grounded inholistic methods and a management framework thatconsiders the complexity and interactions betweenecological, social, and economic systems (e.g.GESAMP 2007). Considering the diverse nature ofbi valve culture techniques and the variable risk of bi -valve aquaculture impacts on marine environments,flexibility within the ecosystem approach is neces-sary. However, regardless of the specific details ofany given aquaculture activity, it is suggested thatthe ecosystem approach address the following mini-mum requirements:

(1) Incorporate the best available scientific knowl-edge of cultured ecosystems, their processes anddynamics and their resulting potential for degrada-tion.

(2) Address potential phytoplankton interactionswith bivalve culture, including effects on suspended

196

Anaerobicsediment

Aerobicsediment

Benthic plants

Harvest

Detritus

Phytoplankton

Biodeposits

Denitrification

Nitrification

NutrientsTidal exchange

Seston

SedimentGrazers

&bacteria

Organic N & PNO3–

NO3–

NO2–

NO2– N2

N2

NH4+

Burial

Mix

ing

N, P

, O2

exch

ange

Run-offSunlight

Bivalves

CB

AD

Fig. 1. Conceptual diagram of bivalve aquaculture interactions in coastal ecosystems related to: (A) the removal of suspendedparticulate matter (seston) during filter feeding; (B) the biodeposition of undigested organic matter in feces and pseudofeces;(C) the excretion of ammonia nitrogen; and (D) the removal of materials (nutrients) in the bivalve harvest (Cranford et al. 2006)


particle dynamics (concentration, composition, sizespectra, transport) and pelagic trophic structure.

(3) Address potential impacts on seabed geophysi-cal properties, geochemical processes, and the struc-ture and ecological role of benthic flora and fauna(i.e. address potential changes to benthic habitat,biodiversity, and ecosystem function).

(4) Address potential interactions between differ-ent farms within the water body and the role of foul-ing communities on culture structures in order toassess the net cumulative environmental effect.

(5) Consider cost versus benefit of implementingdifferent management options.

(6) Consider the potential ecological services pro-vided by the culture activities, including potentialmitigation of eutrophication and increased biodiver-sity.

(7) Integrate ecosystem management with otherrelevant sectors within a management frameworkthat also considers social issues and economicimpacts with respect to setting standards for sustain-ability (i.e. finding an equitable balance that permitsindustry to remain economically viable).

(8) The approach should be inclusive and continu-ous with diverse stakeholder participation, trans-parency, and communication.

Additional general principles of the ecosystemapproach are provided by the Convention on Bio -logical Diversity (www.cbd.int/doc/publications/ea-text-en.pdf).

A management framework that incorporates theecosystem approach must provide the means tostructure sets of measurable indicators of the ecolog-ical, social, and economic factors that are relevant tobivalve aquaculture. This needs to be accomplishedin a way that can address many aspects of a problemin a manner that facilitates their interpretation andaids in the understanding of how different issues areinterrelated. Indicator systems are seen as centraltools for ecosystem-based fisheries management,helping to steer fisheries towards sustainability byproviding timely and useful information to decision-makers (Rice 2003). An indicator is observed or esti-mated data describing a particular characteristic ofthe system. Performance indicators are descriptiveindicators associated with target values (thresholdsor limits). Environmental performance indicators andrelated standards are often presented within alreadyestablished frameworks that are generally built in agiven societal context (Olsen 2003). Several concep-tual management frameworks are reviewed in thefollowing sections to address their potential for adap-tation to sustainable aquaculture development.

Market-driven frameworks for industry self-regulation

A global activity related to the development of anecosystem-based approach for aquaculture is thecreation of performance-based standards that arelinked to certification schemes designed to managethe key social and environmental issues associatedwith bivalve farming. It is believed that the imple-mentation of certification schemes helps the industrysector to work toward more sustainable aquaculture,including reduced impacts. The underlining princi-ple of certification is that a body fully independentfrom the production sector should be responsible forcertification while the costs are borne by industry.Certification schemes relevant in some way to aqua-culture have been reviewed by Funge-Smith et al.(2007) and the World Wildlife Fund (WWF 2007).Organizations active in this field include the Foodand Agri culture Organization of the United Nations(FAO), WWF, Friends of the Sea, Naturland, GlobalGap, and the Aquaculture Certification Council. TheMarine Stewardship Council decided to cease work-ing on aquaculture certification, but continues to be akey participant because it does certify bivalve aqua-culture activities where juveniles are collected fromwild stocks. The WWF Bivalve Aquaculture Dialoguehas recently completed certification criteria and stan-dards for bivalve aquaculture based on performance-based standards. These standards have been givento a new organization (Aquaculture StewardshipCouncil) responsible for certifying farms that are incompliance. The FAO has produced guidelines (FAO2007) intended for the production of improved finfishand bivalve aquaculture certification schemes thatcomply with the main principles of the ecosystemapproach. The FAO specified certification standardsnamed ‘Hazards, Analysis, and Critical ControlPoint’ (HACCP) that are the core of several nationallegislations across Europe and the United States.

A shortcoming in addressing bivalve aquaculturesustainability through a market-driven certificationapproach is that consumer awareness and valuesrelated to environmental impacts vary towards bothextremes across and within geographic markets.Local perceptions on the acceptability of aquacultureimpacts may not match more broadly establishedenvironmental quality criteria enforced by responsi-ble regulatory agencies. Another potential limitationof certification schemes is that they currently do notfully encompass the complexities of interactionsbetween bivalve culture and the ecosystem andtherefore do not meet criteria outlined in legislations

197


that mandate an ecosystem-based approach. Forexample, the HACCP standards do not considersocial and environmental impacts and are not strictlypertinent to the implementation of an ecosystem-based management framework as discussed herein.Third-party certification schemes do not, and are notmeant to, displace an effective governance approachfor ensuring the sustainable use of coastal ecosys-tems. A key benefit to the underlying work that hasgone into the establishment of certification schemesis the compilation of information on societal expecta-tions on the ecological performance of aquacultureoperations. For example, the WWF bivalve certifica-tion standards were developed based on wide stake-holder participation in multiple dialogue workshopsand through open calls for comments on the draft per -formance-based standards. This participatory multi-stakeholder approach, which included science inputat all stages, was an iterative process designed toboth reveal and balance opposing views.

Governance-driven regulatory frameworks

The development and management of aquaculturefalls within the scope of numerous legislations inmany countries, including many that promote anecosystem-based approach to aquaculture manage-ment. Over the past decades, scientists and policy-makers have become increasingly aware of the com-plex and manifold linkages between ecological andhuman systems, which generated a strong researcheffort into social-ecological systems analysis. Social-ecological systems are understood to be complexadaptive systems where social and biophysical agentsare interacting at multiple temporal and spatial scales(Janssen & Ostrom 2006). This has stimulated re -searchers across multiple disciplines to look for newways of understanding and responding to changesand drivers in both systems and their interactions(Zurek & Henrichs 2007). Integrated coastal zonemanagement can be viewed as being part of thissocial-ecological system paradigm, in which specialemphasis is placed on the complexities of coastal set-tings and the various links and drivers in ecologicaland human systems. Using a topic-focused approachto social-ecological systems strengthens this type ofanalysis framework.

Several different types of regulatory frameworkshave evolved within the indicator approach to sus-tainable development and have been applied at theinternational, national, regional, and/or local levels.One management framework that identifies environ-

mental problems, their causes, and solutions, andwhich recognizes important linkages between ecological and socio-economic systems is knownas Drivers-Pressures-State Change-Impact-Response(DPSIR). The DPSIR framework organizes multiplesystems of indicators and is the latest frameworkdeveloped by the Organisation for Economic Co-operation and Development (OECD 2000). An earlyversion of the DPSIR framework, called the Pressure-State-Response (PSR) framework, states that humaninfluences and activities exert pressures on the envi-ronment, which can cause changes in the state (e.g.environmental quality) of the system. Regulatorsthen respond with environmental and economic poli-cies and programs intended to prevent, reduce, ormitigate pressures and/or environmental impact. Thesecond variation added a category of impact indica-tors, transforming it into a Pressure-State-Impact-Response (PSIR) framework. The latest version,which has become widely employed, is the DPSIRframework (Fig. 2A). In this framework, an additionalcategory of indicators describing human influencesand activities that positively or negatively impact theenvironment (driving forces) was added. Each cate-gory within DPSIR is described by a set of indicators,and the framework assumes that all indicators areinter-related. A full description is given in publica-tions by the OECD.

DPSIR is a well developed, science-based frame-work that reveals aspects of environmental prob-lems, their causes, and remedies (Fig. 2B). It allowscoverage of a large spectrum of particular situationsconcerning the environment, including aquaculture,and has been applied globally, including use by theEuropean Environmental Agency to manage the EU‘Water Framework Directive.’ DPSIR is often recom-mended for coastal zone management to identify thekey factors and processes at different stages. Aqua-culture management based on this framework wouldprovide an analysis of aquaculture ecosystem inter-actions in many different types of systems and wouldrecognize critical linkages between ecological andsocioeconomic systems. DPSIR has been criticized forlacking some statistical rigor and for simplifyingcomplex ecological linkages. However, this approachhelps to focus attention on the development of regu-latory-relevant research on each component of theframework and on understanding the linkages. Forexample, numerical models may be developed thatincorporate knowledge of system interactions toexplore the level of complexity needed to describesystem interactions within acceptable confidencelimits (also see below). We therefore recommend the

198


DPSIR framework as a basis for the assessment, eval-uation, and operational management of bivalveaquaculture activities in the coastal zone.

ROLE OF MODELING IN BIVALVE AQUACULTURE MANAGEMENT

Aquaculture impact assessment generally takesplace at the scale of individual farms. Extrapolatingthese effects up to any larger scale is considerablymore complex and includes considering impacts ofmultiple farms and ecological interactions that are afunction of bathymetry, farm proximity, and circula-tion. Modeling permits an understanding of how allfarms interact over a relevant scale and provides ameans to demonstrate our state-of-knowledge ofaquaculture–environment interactions, particularlywhen model output is tested and confirmed. Modelsare essentially a set of mathematical equations thatrepresent particular features of the behavior of bi -valve culture systems and coastal ecosystems. Bivalve

aquaculture models have focused on the con-cept of carrying capacity (see above), andtypes of models vary greatly in complexity(Grant & Filgueira 2011). A relatively unde-manding approach is to estimate and comparethose critical aquaculture processes thatmay cause ecological effects against oceano-graphic processes that may prevent the im-pact from being expressed. For example, afood depletion index is in widespread use thatscales the estimated time it takes for a givenbivalve population to filter a known waterbody with the flushing (residence) time of thatwater body (e.g. Smaal & Prins 1993). This in-dex is included within the 2010 WWF BivalveAquaculture Dialogue Certification Standardsas a practical (e.g. simplest available method)means of assessing the cumulative effects ofall bivalve farms on the ca pa city of a waterbody to support the re gional intensity of culture. This single-box ‘depletion index’ car-rying capacity approach is utilized with theknowledge that it does not provide informa-tion on spatial variability in water flushingand food depletion within a water body.

The ecosystem management domain (spa-tial and temporal boundary scales) and theneed for information on variability at differentscales within this domain are important con-siderations in the design of a managementframework. The possible need to predict sys-

tem status and impact variability at different scalesmay be explored by increasing the level of modelcomplexity. Dynamic models are the most complexmodeling approach and include 2D box models to 3Dfinite element models coupled with hydrodynamicmodels. An advantage of dynamics models is thatthey are able to simulate events, such as changes inan ecological effect over a year. For example, the zoneof potential benthic community effects from bivalvebiodeposits may be predicted using particle trackingmodels, such as DEPOMOD (Cromey et al. 2002,Weise et al. 2009), that predict organic matter flux tothe seabed. The ECASA project has identified avirtual toolbox containing, among other ‘tools,’ a list ofmodels such as ShellSIM, EcoWin, FARM, Longlines,DEB, DEPOMOD, and DDP that can be used by envi-ronmental managers to guide actions that minimizethe environmental im pact from bivalve aquacultureoperations (www. ecasatoolbox.org.uk). These modelsare intended to help maintain environmental qualityand ensure the sustainability of sites and water bodiesfor aquaculture.

199

Drivers

Pressures

State

Impact

Response

Causes

Problem

Solution

Economics/JobsHealth benefits

Ecological servicesPerceptions/Concerns

StructuresGrazing

DefecationExcretion

RespirationHarvestingPathogens

Introductions

Habitat destructionPopulation size/

distributionLoss of biodiversityProduction decline

Ocean policiesConservation/Rehabilitation

Monitoring systemsMitigation measuresEconomic policies

Drivingforces

Pressures

State

Impact

Response

Health/quality of environment

A

B

Fig. 2. The Drivers-Pressures-State Change-Impact-Response (DPSIR)framework for integrating aquaculture, socio-economic, and ecologicalinteractions. (A) Components and functions of the DPSIR framework. (B)Example factors and processes relevant to each component of a DPSIRframework applied towards the management of shellfish aquaculture


Ecosystem models that include bivalve aquaculturescenarios can provide a holistic perspective requiredwithin an ecosystem-based management framework.Models can provide information on potentially criti-cal ecological effects that may be hard to measurebut are suspected to cascade through all trophic lev-els and alter ecosystem structure and function. Sus-pended mussel culture husbandry practices (raft andlong-line) can permit the holding of a relatively highbiomass per unit area and direct access to food sup-plies in the water column. Consequently, modelingactivities have focused, to a large extent, on this formof culture. Grant et al. (2008) used a fully-coupledbiological−physical−chemical model to demonstratethat suspended mussel culture may, under certainconditions, exert a controlling effect on the concen-tration and distribution of phytoplankton and othersuspended particles. Similarly, Cranford et al. (2007)utilized a nitrogen budget and box model to demon-strate how a suspended mussel culture activity con-trolled nutrient dynamics at the coastal ecosystemscale.

Ecosystem models have been used as an integralpart of the DPSIR framework to test scenarios ofaquaculture pressures on water quality (Nobre et al.2005) and ecosystem productivity (Marinov et al.2007). Models are useful to address several compo-nents of the DPSIR framework, and perhaps mostpractically can help to identify performance indica-tors of ecosystem status and aquaculture impacts andtheir associated operational management thresholds.Modeling is also among the few tools capable ofassessing aquaculture sustainability while also con-sidering the cumulative effects of additional humanactivities (e.g. eutrophication, climate change) andresident and invasive suspension-feeding species(Cranford et al. 2007, Ferreira et al. 2008). In sum-mary, it is possible to utilize models to:

(1) assess the potential impact of bivalves on theecosystem state;

(2) define indicators based on predicted ecologicalfluxes that summarize ecosystem functions (e.g.nutrient throughput, recycling, and time scales; Prinset al. 1998);

(3) define ecological thresholds linked to the density-dependant effects of bivalve aquaculture;

(4) compare ecosystems using the selected set ofprognostic indicators;

(5) simulate bivalve culture interactions beforeand/or after the commencement of husbandry, or as aresult of a proposed increase in bivalve standing stock;

(6) optimize the spatial layout of individual farms ina manner that best promotes ecosystem sustainability;

(7) assess interactions between aquaculture andother human activities in the coastal zone; and

(8) assess ecosystem functioning in the long termand determine whether aquaculture ecosystem inter-actions interfere with other services provided by theecosystem.

A number of countries have well-developed poli-cies and procedures in place that utilize modelingtools for planning and monitoring as well as regula-tion of impacts from nutrient enhancement andorganic waste deposition and dispersion (reviewedby Henderson et al. 2001). However, the direct use ofmodels for the regulation and monitoring of aquacul-ture has been restricted to fish aquaculture in a rela-tively small number of countries (Henderson et al.2001). With respect to bivalves, the models that are incurrent use to predict production carrying capacity,food depletion, and ecological interactions have onlybeen indirectly utilized through inclusion of scien-tists in national and regional advisory activities. Practicality issues related to technical complexity inmodel implementation and interpretation and theneed for site-specific parameterization and forcingdata have generally limited the routine use of dy -namic ecosystem modeling for bivalve aquaculturegovernance. However, applications of these modelshave contributed substantially to the state-of-knowl-edge that will be instrumental when developing anoperational ecosystem-based management approach.Model applications have greatly advanced our under -standing of interactions between many componentsof the DPSIR framework.

INDICATORS FOR BIVALVE CULTURE— GENERAL CONSIDERATIONS

Ecological indicators can help to describe ecosys-tem status, ecosystem health, environmental perfor-mance, and functional sustainability performance(Rice 2003, Gibbs 2007). Indicator selection dependson the activity addressed and the spatial or economicscale considered (Spangenberg 2002, Rochet &Trenkel 2003). Indicators therefore need to be identi-fied and developed specifically for bivalve aqua -culture as a way to assess and quantify changes,progress, and improvements towards sustainableindustry development. Ideally, the indicator frame-work for integrative bivalve cultivation assessmentshould transparently encompass the full spectrumof components of the DPSIR framework such thathypotheses about the links between all componentsand management outcomes on the local to national

200


level may be tested. For instance, if managementresponses to observed aquaculture impacts wereeffective, it may result in improved public percep-tions and further aquaculture growth. Failing toincorporate such an auditing component within theframework may promote ineffective policies and prolong the transition to sustainability (Rudd 2004).

No one universal set of indicators is applicable inall cases (Segnestam 2002), and no single indicatorcan account for the whole system; economic, social,and ecological dimensions need to be integrated.However, a small set of well-chosen and highly rele-vant indicators tends to be the favorite choice of mostusers, including the stakeholders for aquaculture.The selection criterion can be applied when there is aneed to constrain the number of indicators. Gilbert & Feenstra (1994) identified 4 desired features of sus-tainability indicators: (1) they must be representativeof the system chosen and must have a scientific basis;(2) they must be quantifiable; (3) a part of the cause-effect chain should be clearly represented; and(4) they should offer implications for policy. These features intersect with the principles of the DPSIRframework. Notwithstanding the need to adapt allcomponents of the DPSIR framework towards aqua-culture application, our focus here is on summarizingand recommending ecological indicators specific tobivalve culture that are related to the ‘state’ and ‘im -pact’ components. Although they specifically addressecosystem components and processes from the farmfootprint to far-field effects (coastal ecosystem scale),these indicators are also of high societal relevance.

The interactions and feedbacks between the eco-nomic, social, and environmental components ofDPSIR play out over time and space. These arereferred to as ‘cross-scale’ or ‘multi-scale’ processes.Indicators for bivalve cultivation commonly focus onprocesses at the geographic scale that influencebivalve aquaculture development, which can includelocal to national decision-making requirements. Theapplication of certain indicators and their respectivedecision thresholds for bivalve cultivation may needto differ between countries and between regions dueto these scale issues and to differences in demands,traditions, cultures, or management systems. To takeaccount of this array of complexity in the context ofdecision-making, a number of research-supportedapproaches to indicator and monitoring systems havebeen developed and advanced to better understandthe current and future interaction of various drivingforces (Carpenter & Brock 2006). Recently, indicatorsystems have also been used to address multi-scaleprocesses or to link social-ecological systems devel-

oped at various geographical scales in order to betterunderstand the interaction of processes, objectives,and institutional arrangements across scales (Car-penter et al. 2008). Such an indicator set would cap-ture environmental status and impacts while alsotracking the state and trends in social satisfactionand economic revenues at different geographic scales.

Indicators need to be relevant to the scale of theissue they address and must address political realtiesif they are to gain acceptance and to achieve practi-cal application. Processes at different geographicscales commonly unfold over different time and spacescales. For example, a system’s social-ecologicaldynamics will unfold more slowly as the spatial scaleexpands from local to regional to national, and morequickly as aquaculture grows over time from conceptto carrying capacity. Indicator approaches may beused to address multi-scale processes and to linkrepercussions at various geographical scales to under -stand more fully the social-ecological dynamics ofbivalve cultivation. Indicators that provide informa-tion that is not scale-dependent will obviously havewide application. However, not all indicators are rel-evant to multiple geographical and temporal scalesbut are needed to provide specialized knowledge ona particular focal issue. It is important that local toregional aquaculture managers understand the con-text in which a selected indicator works.

Several recent papers have proposed a list of per-formance criteria for selecting ecological indicators(Kurtz et al. 2001, Rice et al. 2005) and specificallyfor fishery indicators (Garcia & Staples 2000) andbivalve aquaculture (Cranford et al. 2006, Gibbs2007). The following criteria were used as a guide bythe authors for assessing the potential applicationof various indicators for use in the management ofbivalve culture:

(1) Relevance: The meaning of an indicator repre-sents the first essential phase in the process of indica-tor selection. The indicator selection must be closelylinked to potential negative environmental effectsthat should be addressed.

(2) Effectiveness: This determines the ability ofmanagers to respond to variations in aquacultureforces and pressures. While some indicators mayrespond to dramatic changes in the system, a suitableindicator displays high sensitivity to particular and,perhaps, subtle stress, thereby serving as an ‘early-warning’ indicator of reduced system integrity.

(3) Precision and accuracy: An indicator is consid-ered to be robust if the natural variability of the indi-cator in the environment is low and if the inherentvariability in results from available methodologies

201


and technologies for collecting indicator measure-ments is low. If variability is too high, the samplingdesign (monitoring program) needed to detect anunacceptable effect size may be extensive and im -practical. This is related to the next indicator criteria.

(4) Feasibility: Theoretical indicator constructionsare useless on an operational basis if adequate dataare not available, either due to the fact that the dataare technically very difficult to obtain or if collectingthe necessary information is too expensive. Thus,indicators must be practical and realistic, and theircost of collection and development therefore needs tobe considered. This may lead to trade-offs betweenthe information content of various indicators and thecost of collection. Such pragmatic considerations areparamount for identifying monitoring requirementsfor an industry that includes many small-scale opera-tions. Low-cost measures are obviously preferable ifthey are able to contribute to management objectivesas effectively as more costly approaches.

(5) Sensitivity: A good indicator must be sensitive,with a known substantial response to disturbancesover time.

(6) Clarity: The ease of data interpretation is an im -portant consideration for managers and non-scientistsinvolved in the decision-making process. Indicatorsact as information communication tools and changesin an indicator should easily be understood by stake-holders.

(7) Responsiveness: For management to be effec-tive, the time frame between data collection and thedecision-making process needs to be as short as possible. Responsive and adaptive management ap -proaches strive to implement mitigation measuresquickly so that the impact does not continue toincrease. Near real-time indicators therefore have adistinct advantage in such programs, whereas indi-cators that require considerable work to process samples and interpret data are less desirable. Timelags greater than approximately 6 mo for managersto receive final results for an indicator can be con -sidered detrimental to management.

In the following sections, we assess different indi-cator sets and their feasibility according to the per -formance criteria outlined above. The indicatorsdescribed provide information on ecological systemstatus and, more specifically, the impact of bivalveaquaculture in the coastal zone and were compiledfrom a number of sources. Several European researchcontracts were aimed at producing indicators relatedto the interaction of aquaculture (and bivalve culture)with the marine environment. Examples of attemptsto compile indicators related with the sustainable

development of marine aquaculture include theMARAQUA (Read et al. 2000), Consensus (www.euraquaculture.info/), and ECASA (www.ecasa.org.uk) research programs. A Canadian national sciencead visory review was conducted to identify and rec-ommend an indicator-based ap proach and environ -mental monitoring frame work for managing bivalveaquaculture impacts on marine habitat (DFO 2006,Chamberlain et al. 2006, Cranford et al. 2006). Inaddition, Gibbs (2007) identified sustainability per-formance indicators based on bivalve aquacultureinteractions in the water column. Potential indicatorsare presented according to a scheme that includesbenthic, pelagic, and socio-economic aspects of ma -rine coastal ecosystems. Additional indicators on sitemorphology, hydrography, and husbandry practices(e.g. standing stock) are needed to increase the relia-bility of impact assessments (Borja et al. 2009) andto aid in the interpretation of ecosystem ‘State’ and‘Impact’ indicators.

Benthic habitat indicators

The primary source of bivalve aquaculture effectson seabed habitat (i.e. the properties of sedimentsrequired by a particular organism or population) isthe deposition of excess organic matter in bivalvefeces and pseudofeces and other fall-off from sus-pended culture structures (bivalve and foulingorganisms and their biodeposits). Potential sedimenthabitat state indicators are summarized in Table 1.Some indicators are intended to address the flux oforganic matter to the sediment, some characterizethe change in the sediment properties, while othersdescribe the state of biogeochemical processes asso-ciated with the ecological recycling of depositedorganic matter.

There is a close relationship between observedchanges in all of the indicators presented in Table 1across the full range of the sediment organic enrich-ment gradient (Hargrave et al. 2008, Hargrave 2010).Those studies showed that changes in benthic habi-tat state can be classified based on ‘sulfide regimes.’The concentration of total dissolved (free) sulfide(S2−) in surficial (0−2 cm) sediments is a sensitive,cost-effective, and proven indicator of the organicenrichment effects of bivalve aquaculture on benthichabitat and communities (e.g. Cranford et al. 2006,2009, Hargrave et al. 2008, Hargrave 2010). Aquacul-ture management regimes in Canada, as well as theglobal certification scheme developed by the WWF,focus on total ‘free’ S2− measurements using electro-

202


chemical methods. The collection of data on othergeochemical organic enrichment indicators is alsoencouraged to provide additional habitat protection(confirmation of total dissolved S2− results) and topermit the calculation of composite habitat indices(e.g. Benthic Enrichment Index; Hargrave 1994). Thesediment profile imaging technique offers somepractical benefits, including high sensitivity (e.g.Grant 2010), but after the redox discontinuity layerreaches the sediment surface, visual indicator cuescan appear constant while geochemical conditions

and community impacts continue to degrade withincreasing organic enrichment. Sediment profileimaging has been shown to underestimate seabedorganic enrichment effects (Mulsow et al. 2006) andbe less effective than direct physical and biochemicalquantitative measurements of sediment cores (Kee-gan et al. 2001). Advances in sediment dissolved oxy-gen measurement based on optical methods appearto have overcome past methodological problems thathave limited the practical application of this indica-tor. Benthic biological effects from organic enrich-

203

Indicator Description

Sedimentation rate Sediment trap measurements of the amount (flux of sediment and organic matter)and composition (total organic, carbon, nitrogen, etc.) of particulate matter fallingfrom bivalve culture

Biodeposition rate Quantitative collections of the biodeposits produced by bivalves

Sediment texture Sediment grain size and indices of the increase in the fine fraction (e.g. percentsand-silt-clay)

Organic enrichment Total organic matter and/or organic carbon concentration in surficial sediments

N and P enrichment Total nitrogen, organic nitrogen, and phosphorus concentrations in surficialsediments

Sediment quality Percentage organic carbon, C:N ratio, and refractive proportion (Rp) index (Kristensen 2000)

Redox potential The oxidation-reduction potential (Eh) in surficial sediment is related to theenergy-yielding reactions of bacterial cells (e.g. low Eh values are linked with theanaerobic degradation of biodeposited organic matter). A related indicator is thedepth of the redox potential discontinuity, which represents the depth where Ehpotentials change rapidly from positive to negative values

Total free sulfides Sediment pore water sulfide measurements made using various electrode andspectrophotometric methods. Total S2− in pore water creates toxic biological effectsfor benthic fauna by interference with aerobic respiration

Water content The percentage of surficial sediment dry weight is related to grain size, which canbe altered by addition of bivalve biodeposits

Dissolved oxygen Sediment oxygen concentration measurements. Chemical and biological oxygenuptake increases when organic matter sedimentation increases. Hypoxic andanoxic sediment conditions impact benthic fauna

Benthic/pelagic flux Measurements of the uptake/release of sulfate, oxygen, and nutrients at thesediment/water interface

Pigments Chlorophyll and phaeopigment concentrations measured in surface sediment. Afraction of the phytoplankton ingested by bivalves is not digested and can accu -mulate beneath the facilities

Visual observations Photography and video imaging of the seabed surface and sediment verticalprofiles can reveal changes in sediment color and texture linked to enhancedsediment and organic biodeposition. The presence of sulfur-reducing bacterialmats also indicates reducing conditions at the sediment water interface

Benthic Enrichment Index (BEI) Derived from measurements of surface sediment organic and water content andEh (Hargrave 1994)

Benthic Habitat Quality Index (BHQ) Calculated based on vertical profile imaging of sedimentary structures fromundisturbed sediment (Nilsson & Rosenberg 2000). Vertical profile images ofsediment beneath aquaculture operations show changes in sediment color andorganism distributions indicative of organic enrichment effects

Table 1. Potential indicators of the state of benthic habitat relevant to assessing the effects of increased organic matter deposition by bivalve aquaculture


ment are largely linked to a combination of the toxiceffects of H2S and oxygen deficiency, both of whichcan be measured cheaply, accurately, and rapidly.We therefore suggest a focus on these 2 indicators forcharacterizing the DPSIR benthic ‘State’ parameter.

Benthic community impact indicators

The impacts on benthic communities of increasedorganic matter input to sediments are well known(Pearson & Rosenberg 1978, Hargrave et al. 2008).Hypoxic to anoxic and sulfidic conditions are createdin surface sediments if oxygen consumption ratesexceed the supply, which is limited by physical andbiological factors that control sediment−water ex -change. The macrobenthic community can be ex -pected to exhibit the following responses to anincrease in organic loading:

(1) a decrease in species richness and an increasein the total number of individuals as a result of thehigh densities of a few opportunistic species;

(2) a general reduction in most species biomass,although there may be an increase in total biomasscorresponding to the presence of a few opportunisticspecies;

(3) a decrease in body size of the average speciesor individual;

(4) a shallowing of that portion of the sediment column occupied by infauna; and

(5) a shift in the relative dominance of trophic groups.The choice of community indicators is related to the

specific needs of regulators. Using the DPSIR frame-work as an example, benthic community indica -

tors have often been used as a general indicator ofchanges in ecosystem quality or health (State), butare conceptually suited to monitoring the impact thatresults from changes in benthic habitat. The mostcommonly employed benthic community indicatorsare summarized in Table 2.

Many criteria have been used to select indices ofbenthic community effects, and the range of indicatorsemployed has led to difficulties comparing resultsacross studies. The number and abundance of benthicspecies at a site can be examined in a wide variety ofways to provide an index of biodiversity. H’, j, and c(Table 2) are routinely used to help describe the diver-sity of macrofaunal communities. However, values ob -tained with community diversity and biomass indica-tors should be interpreted cautiously and with a fullunderstanding of what these indicator results actuallyreveal about community changes. Consequently, com -munity impact programs typically require the use ofseveral indicators to interpret changes in the faunaand the probable cause of biodiversity change in as-sessing farm impacts (e.g. Borja et al. 2009). Multi-variate analysis is a powerful tool for identifying pat-terns in the large amount of data on species andabundance produced from a site survey. It can also in-corporate other data on environmental conditions atthe site to provide an understanding of the ecologyand environmental degradation of the area under study.

Information on opportunistic and scavenger species(i.e. indicator species), which tend to increase in num-ber under some bivalve farms, is relatively easily in-terpreted compared to results based on diversity in-dices. Deposit-feeding polychaete taxa and largecarnivorous nematode worms have been observed to

204

Indicator group Indicators and methods

Biodiversity metrics Including the Shannon-Wiener diversity index (H’ ), Pielou’s evenness index ( j ), Simpson’sdominance index (c), and Margalef’s species richness (d ). Generally used for macrofauna,but can be applied to assess meiofauna diversity

Indicator species Highly enriched marine sediments are generally dominated by a few opportunistic macro-faunal species, such as Capitella sp., that are tolerant of high organic enrichment and lowoxygen conditions. The AZTI Marine Biotic Index is calculated based on the relative propor-tion of 5 species groups (previously classed as being sensitive to opportunistic)

Trophic indices In highly organically enriched areas, benthic communities are dominated by deposit feedersand scavengers, at the expense of filter feeders. The Infaunal Trophic Index provides acategorization of overall species abundance within different trophic groups in soft bottomcommunities

Benthic similarity Comparison of community structure with multivariate statistics

Size structure Most species that are tolerant to organic enrichment belong to families such as the Spi-onidae, and have a small size. Differential sieving of sediment for macrofauna studies, on1 mm and 0.5 mm sieves, allows quantification of the relative contribution of the smallerindividuals to the whole community

Table 2. Indicators of the intensity of benthic community impacts from organic matter deposition by bivalve aquaculture


dominate in organically enriched sites where Mol-lusca and Echinodermata are completely excluded.The reduction/absence of suspension feeders and in-crease in deposit feeders are also good indicators oforganic matter perturbation. The loss of bio-irrigatingspecies (deep-burrowing infauna) is ecologically im-portant because this could further enhance anoxicconditions caused by organic enrichment. The AZTIMarine Biotic Index (AMBI; Muxika et al. 2005) incor-porates these indicator species concepts and providesa pollution classification for a given site that representsbenthic community ‘health.’ The application of AMBIto a given site requires the availability or developmentof an impact tolerance list and produces a number in arange from 0 to 7. While this indicator is not specific toan aquaculture impact, it reacts predictably to thepresence of organic enrichment (Muxika et al. 2005,Callier et al. 2008). Studies on the effects of bivalvefarming confirm the predicted transition in AMBIbenthic status with increasing organic biodepositionunder mussel lines in Canada (Weise et al. 2009) andalso suggest a decrease in benthic impacts with dis-tance from farms in Europe (Borja et al. 2009).

The Infaunal Trophic Index (ITI) takes a functionalapproach to impact assessment and is based on thefeeding types of the fauna encountered at the surveysites. Benthic similarity analysis can determine differ-ences between sites even at low organic enrichmentand provides a reliable indication of impacts beencontrol and farm sites, and along transects leading

away from farms. ITI and AMBI are proven ap-proaches for indicating impacts and are highly recom-mended for monitoring benthic community impacts.All approaches that require taxonomic identificationare expensive to conduct, and analysis/interpretationtimes can be extensive. It has been suggested that thestrong correlations that exist between observationsof the degree of benthic community impacts and sur -ficial sediment geochemical changes permit geo-chemical indicators to serve as low-cost proxies ofcommunity effects (Hargrave et al. 2008).

Pelagic indicators

Farm structures may affect water flow both withinfarms (Strohmeier et al. 2005) and over the scale ofcoastal ecosystems (Plew 2011) by causing drag.These structures also generate turbulence, which canmix the water column or distort stratification (Stevens& Petersen 2011). Bivalve aquaculture may also altera number of chemical and biological properties in thewater column. Owing to the movement of the water,these effects can be transported far-field with thepotential for a measurable impact at the coastalecosystem scale (Cranford et al. 2006, 2008, Gibbs2007). Several pelagic indicators have been proposedto monitor potential changes in environmental qual-ity (State) and lower trophic levels (Impact) related tobivalve culture (Table 3).

205

Indicator Description and methods

Nutrient concentration There is ample evidence to link bivalve aquaculture as a major control on coastal nutrientdynamics

Dissolved oxygen The biological oxygen demand within the water body is increased by the respiration of the concentration culture, the fouling community, and the remineralization of organic wastes

Bacterial abundance A greater abundance of naturally occurring bacteria may occur due to availability oforganic matter in bivalve biodeposits and the consumption of some fraction of the naturalmicroplanktonic grazer community by the cultured bivalves

Phytoplankton biomass, The natural high variability in phytoplankton and seston biomass (chlorophyll a and/orseston concentration, and suspended particulate matter concentration) requires extensive, high-resolution synoptic particle depletion surveys to detect the magnitude and scale of food depletion at the farm to bay scale

Depletion index This bay-scale indicator is not directly measured, but is computed using site-appropriatedata on the bivalve culture, their feeding rates, and the flushing characteristics of thewater body (e.g. Prins et al. 1998)

Phytoplankton size The picoplankton contribution index (proportion of total phytoplankton) indicates anylarge-scale changes in the size structure of the microbial plankton community that canresult from removal of larger phytoplankton by culture bivalves. Analysis is based on size-fractioned chlorophyll a analysis (Cranford et al. 2006)

Trophic heterogeneity Spatial and temporal scales of variability in the availability of suspended food sourcesmay be generated by bivalve-mediated food depletion. This heterogeneity may bedetected through trophic studies using stable isotopes of C and N in bivalve tissues(Lefebvre et al. 2009)

Table 3. Indicators of the intensity of pelagic alterations and impacts from the activities of cultured bivalves


Although there is ample evidence to show thatbivalve aquaculture can be a major control oncoastal nutrient dynamics and phytoplankton bio-mass, the use of nutrient and chlorophyll a (chl a)concentrations as impact indicators is challenging.This is attributed to the high natural short- to long-term variability in these indicators in coastal sys-tems. In many cases, a single chemical measure-ment or abundance does not prove to be aneffective indicator, whereas a statistic computedfrom data in an exposed site will better reflect thephenomenon. For example, changes in differentnutrient ratios may be more informative than con-centration measurements and may act as suitableproxies for detecting impacts on nutrient dynamics.Similarly, a few isolated measurements of chl aare not relevant for indicating the magnitude andextent of food reduction resulting from bivalve feed-ing. However, synoptic surveys with towed sensorvehicles have provided intensive information onaquaculture-related changes in suspended particledistributions (phytoplankton biomass and total sus-pended particle concentrations) around suspendedbivalve farms (Cranford et al. 2006, Gibbs 2007).These results provide details on the extent (horizon-tal and vertical) and magnitude of phytoplanktondepletion by the bivalve culture, which contributeto direct and model-based assessments of bivalvecarrying capacity (e.g. Grant et al. 2008). Concomi-tant with any significant phytoplankton depletionwill be a similar reduction in the zooplankton viadirect consumption and possibly from reduced foodavailability. Although high-resolution environmentalsensor surveys are rapid and the data are valued,the high capital and technical costs related to usingassociated indicators presently limits their applica-tion for routine monitoring.

Isotopic signatures in consumers of pelagic foodresources can reflect changes in the quality andquantity of suspended organic matter, predator/preyinteractions, and biogeochemical cycles. To ourknowledge, no studies have measured trophic stateresponses to bivalve aquaculture, but sessile organ-isms such as bivalves are promising indicator speciesfor this type of analysis (Lefebvre et al. 2009). Apotential consequence of large-scale phytoplanktondepletion by large bivalve populations is an increasein the proportion of small phytoplankton. Bivalves donot effectively retain picophytoplankton during feed-ing, and this selective feeding behavior may result inbay-scale shifts in size distribution at the base of thefood web, particularly in poorly flushed aquacultureembayments (Cranford et al. 2006, 2008). This shift

towards smaller phytoplankton is also aided bybivalves consuming predators that naturally controlpicophytoplankton abundance. Significant changesin phytoplankton size structure and abundance havethe potential to cause cascading effects through alltrophic levels owing to changes in predator/preyinteractions. Consequently, indicators of phytoplank-ton size (e.g. relative concentrations of chl a retainedon 0.2 and 3 µm pore size filters) are perceived asbeing highly beneficial for use in ecosystem-basedmonitoring programs in extensively farmed bivalveaquaculture inlets. This recommendation is alsolinked to the relatively low cost of performing size-fractioned chl a analysis and the fact that site-specificmeasurements of plankton community alterationsgenerally reflect conditions over large spatial andtemporal scales of impact.

Bivalve performance indicators

Potential bivalve performance indicators includegrowth rate, condition index, and meat yield per hus-bandry effort. These measures do not reveal informa-tion on specific changes in the structure and func-tioning of ecosystems, but provide an indication as towhether bivalve aquaculture is affecting the systemto a greater extent than can be absorbed by naturalprocesses (i.e. environmental status indicators). Inthis respect, particle depletion and bivalve perfor-mance measurements are highly complementary.The former indicator provides information onchanges in pelagic food supplies that likely controlthe latter (see Rosland et al. 2011). Monitoringbivalve condition is a way to assess how the physio-logical status of the bivalve is affected by food limita-tion. Bivalve performance indicators may be mea-sured using cultured and separately caged animals.The strength of caging bivalves for this purpose isthat standardization of performance measures is sim-plified. However, in the typical case of large spatial(horizontal and vertical) and temporal variability innatural environmental conditions that control bivalveperformance in the farmed region (e.g. temperature,currents, food abundance and nutritional quality,and salinity), caged bivalve indicator results arehighly site-specific. This leads to difficulties in extra -polating monitoring data to the scale of aquacultureoperations.

Long-term trends in total bivalve production of awater body have been used to assess the effects ofincreasing stocking density on bay-wide aquacultureproduction, presumably due to depletion of natural

206


food supplies (Héral et al. 1989, Cranford et al. 2010).Time-series measurements of aquaculture produc-tion are therefore useful indicators of impacts on par-ticulate food supplies, including the phytoplankton.These data are generally collected by aquacultureoperations for purposes other than use as a generalindicator of bay-scale ecological status. A majorproblem with all types of bivalve performance indi-cators is that setting of a maximum bivalve perfor-mance threshold for a specific site requires produc-tion carrying capacity to be exceeded (point where areduction in the performance indicator values isobserved). This is not a desirable trait for any sustain-ability indicator. A more important application offarm stocking and production data is that they arerequired in the interpretation of farm impactsobserved using other indicators.

SOCIO-ECONOMIC CONSIDERATIONS ANDINDICATORS

An integrated ecosystem-based bivalve aquacul-ture management approach requires assessing socio-economic dynamics (current status and impacts) aswell as the environmental dimensions of sustainabil-ity. Indicators of socio-economic issues not only needto measure the financial operating performance ofcommercial bivalve farms but also the wider impactsof aquaculture on society at large. Indeed, it is pre-cisely these impacts which, within the DPSIR frame-work, can be expected to invoke a governanceresponse intended to alter the way in which aquacul-ture is regulated and managed. Among the many dif-ferent indicators proposed in the literature, some areof direct relevance for bivalve culture operations.They are related to 4 different overarching socialdimensions, namely (1) the social acceptability of thebivalve culture, (2) the supply availability to the mar-ket, (3) the livelihood security for the local communi-ties, and (4) the economic efficiency of bivalve cul-ture operations.

Bivalve culture may cause visual intrusion, whichmay impact tourism, or it may compete for space withother coastal activities in a spatially constrained envi-ronment. These can be evaluated by means of obser-vations and regular interviews with local stakehold-ers and government bodies. The social acceptabilityof the bivalve culture operations may be evaluatedby means of regular enquiries, using statistical treat-ments of the public attitude towards aquaculture(Whitmarsh & Wattage 2006) and/or assessment ofemerging and existing conflicts. However, emotional

ownership of the sea/coastal area by the local resi-dents/stakeholders and the social values that drivethese ownerships are difficult to capture. ‘Contract-ing costs’ of getting a group of people to agree on anissue (not necessarily in money) are difficult to assessgiven that bivalve aquaculture is developed indiverse coastal settings and may be contested andnegotiated by individuals and communities accord-ing to local social practices, economic conditions, andenvironmental perceptions.

Indicators of cultured bivalve availability to themarket (supply) correspond, to some extent, withconsumption statistics that are usually computed atnational levels. Indicators of livelihood security forthe local communities, such as the tax revenuesfrom leasing plots, correspond to the well-being ofthe bivalve producer on the local level. Indicatorsthat address this issue pertain to income per capitaand employment rate. The importance of aquacul-ture in supporting local livelihoods is most directlymeasured by per capita income in this sector. Aproxy measure may be derived based on the ratio ofgross value added to employment. Total employ-ment is a measure of the scale or ‘importance’ of theaquaculture industry in absolute terms. This is anindicator of the number of people who are depen-dent on aquaculture directly (and indirectly) fortheir livelihood. It has a political as well as an eco-nomic significance.

Indeed, one of the most important groups of indi-cators relates to the direct economic efficiency of aparticular bivalve aquaculture operation. These canbe gauged from productivity ratios, protection costs,and profits. Productivity is a measure of output perunit of input. For instance, trends in labor pro -ductivity are an important indicator of technicalprogress in aquaculture, and productivity differ-ences between farms may indicate which farms aremost vulnerable to falling prices and profits. Protec-tion costs may be incurred in dealing with the envi-ronmental impacts of aquaculture. These likely con-sist of compliance costs incurred by bivalve farmsarising from the obligation of farms to undertakeimpact assessments and regulation, surveillance, andenforcement costs by the respective institutions. En -vironmental protection costs are the counterpart ofenvironmental damage costs. Thus, an inverse re -lationship between these can be expected. Profi -tability is a basic indicator of financial viability. Inthe absence of published data, the profitability of abivalve operation can be addressed and calculatedfrom its component elements (i.e. input costs, pric-ing of products, etc.).

207


OPERATIONAL THRESHOLDS OF POTENTIALCONCERN

Operational thresholds define the targets that needto be addressed by managers and regulators in anyaquaculture monitoring program. ‘Threshold’ is ageneral term of value that can be determined byadministrative and advisory scientific processes. Forexample, if a strict policy was mandated, a thresholdmay be ‘no change in benthic habitat from organicenrichment.’ That is a threshold derived from policyimplementation of a sense of what is socially accept-able. The operational expression might be ‘no morethan 1500 µM of total free sulfide.’ The threshold inthis case is set by a policy statement. In contrast, ifthe desire is to prevent mortality of a specific benthicorganism, one might set a threshold defined by sci-entific knowledge of the organisms’ response to apredicted environmental change. There are otherless well-defined thresholds which describe the pointat which ecosystems show a sudden regime shiftfrom one state to another (e.g. benthic assimilativecapacity). In identifying a threshold, it is important tobe clear on whether the threshold is determined bypolicy decisions or by changes in ecosystems.

In the case of large areas of bivalve cultivation, it isdifficult to set scientific thresholds as there is consid-erable spatial and temporal variability in the naturaldistribution of water and sediment quality parame-ters. The use of thresholds is often based on meanvalues or another measure of central tendency. How-ever, the ecosystem’s response to a disturbance maybe an increase in variability and it may be possible toobserve no change in the mean values of the indiceseven though the variability may increase (e.g. War-wick & Clarke 1993). This is important because it isoften the occasional extreme in environmental vari-ability that shifts ecological status. Consequently, it isdifficult to set a threshold, and sometimes the crite-rion is simply a ‘no net loss’ or ‘no change.’ To set anadequate threshold, scientists, managers, and allstakeholders must together identify the value of ac -ceptable change from reference conditions. To addressthese difficulties, ecosystem managers increasinglyuse a monitoring endpoint, known as thresholds ofpotential concern (TPC), to decide when manage-ment intervention is needed (Biggs & Rogers 2003).TPCs are a set of operational goals along a contin-uum of change in selected environmental indicators(Gillson & Duffin 2007). TPCs are being continuallyadjusted in response to the emergence of new eco-logical information or changing management goals.They provide a conceptual tool that enables ecosys-

tem managers to apply variability concepts in theirmanagement plans, by distinguishing normal ‘back-ground’ variability from an important change ordegradation (Gillson & Duffin 2007).

PERFORMANCE-BASED MANAGEMENTAPPROACH FOR BIVALVE AQUACULTURE

Evaluation of impacts specific to bivalve aquacul-ture presents a particular challenge owing to thewide range of culture species, husbandry practices,environmental settings, and variable spatial scale.Bivalve aquaculture farms may range from less than0.5 ha to operations in some regions that include alarge fraction of total coastal embayment volume.Given the highly diverse nature of the bivalve aqua-culture industry, it is not sufficient to simply providea toolbox of potential indicators of environmental sta-tus specific to bivalve aquaculture, it is equally im -portant to make recommendations, based on soundscience, as to which tools are most appropriate underdifferent conditions.

Aquaculture monitoring has generally focused onthe benthic marine habitat in the immediate vicinityof a farm. This stems from the fact that the seabedhabitat and species composition reflect the synergis-tic effects of past and present activities as well as nat-ural processes that assimilate or disperse particulatewastes. There is also a relatively high level of under-standing of benthic organic enrichment effects andthe interconnectedness between the various indi -cators. Fortunately, this knowledge has alreadyresulted in identification of scientifically defensibleindicator/threshold classification schemes. Examplesinclude the total ‘free’ sulfides classification (Har-grave et al. 2008) and the AMBI index. Effectivemeasures are also available for mitigating benthicorganic enrichment impacts, and prescribed responsesto environmental degradation could be linked tooperational thresholds defined within a responsiveecosystem-based management framework.

It is not sufficient to rely completely on local ben-thic geochemical and community parameters, as theydo not necessarily encompass effects at the ecosys-tem level. Some combination of modeling and incor-poration of additional indicators that target potentialpelagic effects and higher trophic levels are neededover relatively large (inlet-scale) areas to adequatelyassess bivalve aquaculture effects. Furthermore,these should be accompanied by socio-economicindicators. The inability to adequately define quanti-tative operational thresholds for many highly rele-

208


vant indicators of ecosystem status (particularly thosedescribing the structure and dynamics of the watercolumn), owing to present gaps in our knowledge ofecosystems, should not preclude their potential use.Surveillance sampling programs based on selectedwater column parameters are recommended underconditions where environmental impact assessmentsand ongoing monitoring data indicate a relativelyhigh risk of bay-scale impacts. Of particular concernare potential impacts on suspended particle concen-trations and the resulting alterations in pelagicmicroflora and fauna communities and cascadingeffects through the pelagic food web. Surveillancemonitoring of a suite of ecosystem traits that arethought to affect productivity, community structure,and habitat (i.e. contextual indicators; Gavaris et al.2005), is highly warranted when and where signifi-cant particle depletion by bivalve aquaculture is pre-dicted. Seston depletion modeling capabilities haverapidly progressed in recent years and include somerelatively simple quantitative assessment approachesand decision support systems. Surveillance of pelagicindicators would complement benthic operationalmonitoring and would support the basic monitoringprinciple of delineating cause−effect relationships.

Bivalve aquaculture management needs to incor-porate a high degree of flexibility owing to the highdiversity of this industry. It is therefore recom-mended that environmental assessments be based ona tiered indicator monitoring approach that is struc-tured on the principle that increased environmentalrisk requires an increase in monitoring effort. Thisprinciple is employed in the Norwegian and Cana-dian management of finfish aquaculture impact onthe benthic environment (Hansen et al. 2001, Stige-brandt 2011). Various levels of monitoring could betriggered based on:

(1) the nature of the operation (e.g. species, bottomversus suspended culture, proximity to other farms,and stocking density per area or volume);

(2) the environmental risk predicted during aqua-culture site assessments (e.g. ecological risk assess-ments and model-based predictions);

(3) the ongoing measurement of environmentalindicators towards verification of operational thresh-olds; and

(4) other environmental sensitivity indices (e.g.presence of valued ecosystem components and sensi-tivity designations).

In addition, instead of partitioning the range ofvariation of each indicator into 2 classes (acceptableversus unacceptable), a few more threshold classesmay be more pertinent as these can be linked to

management responses/mitigations that allow actionsto be taken that prevent more critical thresholdsbeing exceeded in the future. A tiered responsivemanagement framework, based on sediment geo-chemical classification, is currently employed inparts of eastern Canada to manage benthic effectsfrom aquaculture and has been adapted to someextent in the 2010 WWF Bivalve Aquaculture Dia-logue standards (WWF 2010). Progressively morerigorous monitoring and regulatory responses areautomatically implemented in response to degradingsite classification.

Under the recommended responsive managementframework, decision-makers would have the abilityto increase or decrease the level of monitoring re -quired for any site during subsequent samplingcycles, based, at least in part, on the review of allmonitoring program results. Prior to implementingany monitoring program, it is important that baselinedata on environmental status be collected so thatdecisions on the appropriate level of monitoring canbe made.

TIERED MONITORING APPROACH

For bivalve aquaculture farms that have beenassessed as having a relatively low risk, only a mini-mal level of monitoring appears to be warranted.This monitoring level is intended to be a rapidscreening method for periodic evaluations of bivalveaquaculture impacts at a given site. Low cost, rapidscreening indicators that may be appropriate for thislevel of site assessment include the collection of benthic images (still photographs and/or video) andsimple model estimations (depletion index approach)of the potential for bivalves on this site to contributeto large-scale phytoplankton depletion within thewater body (Table 2). The primary operationalthresh old that could be addressed in this level ofmonitoring is the appearance of Beggiatoa mats (bac-teria that oxidize H2S as an energy source) on theseabed, that are indicative of a relatively high degreeof organic enrichment and a hypoxic to anoxic sedi-ment classification. As with all types of monitoring,the choice of reference sites for comparison with farmdata would include consideration of general bathy-metric, hydrographic, and seabed type conditions inboth areas.

A second monitoring tier is recommended to pro-vide more frequent information on environmentalstatus and impacts at sites where there are indica-tions (predictions) or previous measurements of

209


organically enriched seabed conditions known to bedeleterious to marine organisms. Possible indicatorrecommendations for this secondary monitoring tierwould include a quantitative assessment of surfacesediment (e.g. upper 2 cm) geochemical conditions(redox potential, total ‘free’ sulfides, organic content,and water content). Sampling would be best con-ducted annually in late summer/early fall when thebiological oxygen demand of surface sediments isgreatest. Discrete water sampling could target bay-scale changes in the size structure of the phytoplank-ton (picophytoplankton proportion). An annualbivalve inventory report for the farm would aid ininterpreting indicator results, as it would for all mon-itoring tiers.

The most intensive monitoring tier is recom-mended for assessing sites that are predicted to pre-sent a relatively high environmental risk and/or theresults of previous monitoring at the site showdegrading habitat status. In addition to conductingall components of the above programs, the annualcollection of data could be expanded to include ben-thic community analysis (AMBI, ITI, etc.). Benthicsampling effort should also be increased to givegreater spatial delineation of the impacted area. Aswith the other monitoring tiers, it is recommendedthat regulators have the capacity to alter the level ofmonitoring required for any site during subsequentsampling cycles.

The recommended bivalve aquaculture manage-ment approach attempts to provide sufficient flexibil-ity to be adapted to a diverse array of bivalve hus-bandry methods, cultured species, and farmingintensities. The tiered indicator toolbox and multipledecision threshold approach for assessing environ-mental changes and degradation (DPSIR Statechanges and Impacts) is intended to permit regula-tors the capacity to adapt the approach dependingon site-specific differences in environmental risks,degree of impact, and regional social values on sus-tainability issues. It also provides a means of imple-menting set impact mitigation measures at an earlystage of observed environmental degradation. Eco-nomic considerations have also been addressed tosome extent by describing environmental monitoringtiers that progress from the use of low-cost, semi-quantitative indicators, to more intensive monitoringand surveillance programs that can be selectedbased on the level of environmental risk. Inherentwithin recommending a management approach isadherence to the basic principle that monitoring andmanagement programs be continually adaptive tochanges in our state of knowledge concerning poten-

tial environmental impacts, indicators, and relatedmethodological approaches.

The basic purpose of the DPSIR framework is toensure that any significant adverse effects of bivalveaquaculture pressures on environmental quality andrelated societal benefits (State change and Impacts)will trigger an appropriate response to ensure thatthe economic and societal benefits of aquaculturecan be delivered in a sustainable manner. It is criticalto maintain broad stakeholder communication atall steps in the development and utilization of thisframe work. Societal perceptions of aquaculture im -pacts and sustainability, and regulatory competenceoften represent a critical negative driver for aquacul-ture. Closing the DPSIR loop between regulatoryresponses and aquaculture drivers (Fig. 1) is criticalfor the future expansion of this industry. This linkagecan best be maintained through the establishment ofa transparent and well-communicated managementframework that includes stakeholder participation atall stages and which promotes a broad awareness ofthe consequences and effectiveness of regulatorydecisions.

Acknowledgements. We thank those members of the ICESWorking Group on Shellfish Culture (WGMASC) that orga-nized and hosted annual group meetings. T. Sephton, pastchair of the ICES Mariculture Committee, was instrumentalin encouraging an expansion of management advisory activ-ities within the WGMASC that eventually led to the prepa-ration of this document. Earlier drafts of this paper benefitedfrom constructive comments by A. Bodoy, E. Black, and 3anonymous reviewers.

LITERATURE CITED

Biggs HC, Rogers KH (2003) An adaptive system to link sci-ence, monitoring and management in practice. In: duToit JT, Rogers KH, Biggs HC (eds) The Kruger experi-ence: ecology and management of Savannah hetero-geneity. Island Press, Washington, DC, p 59−80

Borja Á, Rodríguez J, Black K, Bodoy A and others (2009)Assessing the suitability of a range of benthic indices inthe evaluation of environmental impact of fin and shell-fish aquaculture located in sites across Europe. Aqua -culture 293: 231−240

Callier MD, McKindsey CW, Desrosiers G (2008) Evaluationof indicators used to detect mussel farm influence onthe benthos: two case studies in the Magdalen Islands,Eastern Canada. Aquaculture 278: 77−88

Carpenter SR, Brock WA (2006) Rising variance: a leadingindicator of ecological transition. Ecol Lett 9: 311−318

Carpenter SR, Brock WA, Cole JJ, Kitchell JF, Pace ML(2008) Leading indicators of trophic cascades. Ecol Lett11: 128−138

Chamberlain J, Weise AM, Dowd M, Grant J (2006) Model-ing approaches to assess the potential effects of shellfishaquaculture on the marine environment. DFO Can Sci

210


Advis Sec Res Doc 2006/032. Fisheries and OceansCanada, Ottawa. Available at www.dfo-mpo.gc.ca/ csas/Csas/DocREC/2006/RES2006_032_e.pdf

Christensen PB, Glud RN, Dalsgaard T, Gillespie P (2003)Impacts of longline mussel farming on oxygen and nitro-gen dynamics and biological communities of coastal sed-iments. Aquaculture 218: 567−588

Cloern JE (1982) Does the benthos control phytoplanktonbiomass in south San Francisco Bay? Mar Ecol Prog Ser9: 191−202

Cohen LD, Dumbauld BR, Judge ML (2011) Expandingshellfish aquaculture: a review of the ecological servicesprovided by and impacts of native and cultured bivalvesin shellfish-dominated ecosystems. In: Shumway SE (ed)Aquaculture and the environment. Wiley-Blackwell,Chichester, p 239−295

Cranford PJ, Anderson R, Archambault P, Balch T and oth-ers (2006) Indicators and thresholds for use in assessingshellfish aquaculture impacts on fish habitat. CSAS-DFORes Doc 2006/034. Fisheries and Oceans Canada,Ottawa. Available at www.dfo-mpo. gc. ca/ CSAS/ Csas/DocREC/2006/RES2006_034_e.pdf

Cranford PJ, Strain PM, Dowd M, Hargrave BT, Grant J,Archambault MC (2007) Influence of mussel aquacultureon nitrogen dynamics in a nutrient enriched coastalembayment. Mar Ecol Prog Ser 347: 61−78

Cranford PJ, Li W, Strand Ø, Strohmeier T (2008) Phyto-plankton depletion by mussel aquaculture: high res -olution mapping, ecosystem modeling and potential indi cators of ecological carrying capacity. ICES CM2008/H:12. International Council for the Exploration ofthe Sea, Copenhagen. Available at www.ices.dk/products/CMdocs/CM-2008/H/H1208.pdf

Cranford PJ, Hargrave BT, Doucette LI (2009) Benthicorganic enrichment from suspended mussel (Mytilusedulis) culture in Prince Edward Island, Canada. Aqua-culture 292: 189−196

Cranford P, Balch T, Buchan C, Crocker J and others (2010)Mussel aquaculture regulatory effectiveness monitoring: validation of the St. Ann’s Harbour EnvironmentalAssessment and Monitoring Program. DFO Can TechRep Fish Aquat Sci 2905. Fisheries and Oceans Canada,Ottawa. Available at www.dfo-mpo.gc.ca/ Library/ 343527.pdf

Cromey CJ, Nickell TD, Black KD (2002) DEPOMOD-mod-elling the deposition and biological effects of wastesolids from marine cage farms. Aquaculture 214: 211−239

D’Amours O, Archambault P, McKindsey CW, Johnson LE(2008) Local enhancement of epibenthic macrofauna byaquaculture activities. Mar Ecol Prog Ser 371: 73−84

Dame RF (1996) Ecology of marine bivalves: an ecosystemapproach. CRC Press, Boca Raton, FL

DFO (Department of Fisheries and Oceans Canada) (2006)Assessing habitat risks associated with bivalve aquacul-ture in the marine environment. DFO Can Sci Advis SecSci Advis Rep 2006/005. Fisheries and Oceans Canada,Ottawa. Fisheries and Oceans Canada, Ottawa. Avail-able at www.dfo-mpo. gc.ca/csas/ Csas/ status/ 2006/SAR-AS2006_005_E.pdf

Dumbauld BR, Ruesink JL, Rumrill SS (2009) The ecologicalrole of bivalve shellfish aquaculture in the estuarineenvironment: a review with application to oyster andclam culture in the West Coast (USA) estuaries. Aqua -culture 290: 196−223

FAO (Food and Agriculture Organization of the United

Nations) (2007) FAO technical guidelines on aquaculturecertification. FAO, Rome. Available at ftp: /ftp.fao.org/docrep/ fao/ meeting/014/ai770e.pdf

Ferreira JG, Hawkins AJS, Monteiro P, Moore H and others(2008) Integrated assessment of ecosystem-scale carry-ing capacity in shellfish growing areas. Aquaculture 275: 138−151

Forrest BM, Keeley NB, Hopkins GA, Webb SC, ClementDM (2009) Bivalve aquaculture in estuaries: review andsynthesis of oyster cultivation effects. Aquaculture 298: 1−15

Funge-Smith S, Corsin F, Clausen J (2007) Overview ofaquaculture certification. Expert Workshop on Guide-lines for Aquaculture Certification, 27–30 March 2007,Bangkok, Thailand. Available at www.enaca.org/mod-ules/certificationprojects/index.php?content_id=3

Garcia SM, Staples DJ (2000) Sustainability reference sys-tems and indicators for responsible marine capture fisheries: a review of concepts and elements for a setof guidelines. Mar Freshw Res 51: 535−541

Gavaris S, Porter JM, Stephenson RL, Robert G, Pezzack DS(2005) Review of management plan conservation strate-gies for Canadian fisheries on Georges Bank: a test ofa practical ecosystem-based framework. ICES CM2005/BB: 05. International Council for the Exploration ofthe Sea, Copenhagen

GESAMP (IMO/ FAO/ UNESCO-IOC/WMO/WHO/ IAEA/UN/ UNEP Joint Group of Experts on the ScientificAspects of Marine Environmental Protection) (2001)Planning and management for sustainable coastal aqua-culture development. GESAMP Rep Stud 68. FAO, Rome

Gibbs MT (2007) Sustainability performance indicators forsuspended bivalve aquaculture activities. Ecol Indic 7: 94−107

Gilbert AJ, Feenstra JF (1994) A sustainability indicator forthe Dutch environmental policy theme ‘Diffusion’: cad-mium accumulation in soil. Ecol Econ 9: 253−265

Gillson L, Duffin KI (2007) Thresholds of potential concernas benchmarks in the management of African savan-nahs. Philos Trans R Soc B Biol Sci 362: 309−319

Grant J (2010) Coastal communities, participatory research,and far-field effects of aquaculture. Aquacult EnvironInteract 1: 85−93

Grant J, Filgueira R (2011) The application of dynamic modelling to prediction of production carrying capacityin shellfish farming. In: Shumway SE (ed) Aquacul -ture and the environment. Wiley-Blackwell, Chichester,p 135−154

Grant J, Bacher C, Cranford PJ, Guyondet T, Carreau M(2008) A spatially explicit ecosystem model of sestondepletion in dense mussel culture. J Mar Syst 73: 155−168

Hansen PK, Ervik A, Schaaning M, Johannessen P, Aure J,Jahnsen T, Stigebrandt A (2001) Regulating the localenvironmental impact of intensive marine fish farming II.The monitoring programme of the MOM system (Model-ling - Ongrowing fish farms - Monitoring). Aquaculture194: 75−92

Hargrave BT (1994) A benthic enrichment index. Can TechRep Fish Aquat Sci 1949: 79−91

Hargrave BT (2010) Empirical relationships describing ben-thic impacts of salmon aquaculture. Aquacult EnvironInteract 1: 33−46

Hargrave BT, Holmer M, Newcombe CP (2008) Towards aclassification of organic enrichment in marine sedimentsbased on biogeochemical indicators. Mar Pollut Bull 56:

211


810−824Henderson AR, Gamito S, Karakassis I, Maroni K, Pedersen

PB, Smaal AC (2001) Use of hydrodynamic and benthicmodels for managing environmental impacts. J ApplIchthyol 17: 163−172

Héral M, Bacher C, Deslous-Paoli JM (1989) La capacité bio-tique des bassins ostréicoles. In: Ifremer (ed) L’homme etles ressources halieutiques. IFREMER, Brest, p 225−259.Available at http: //archimer.ifremer. fr/ doc/ 00000/ 3011/(accessed on 1 May 2012)

Janssen MA, Ostrom E (2006) Governing social-ecologicalsystems. In: Tesfatsion L, Judd KL (eds) Handbook ofcomputational economics, Vol 2. Elsevier, Amsterdam,p 1465−1509

Keegan BF, Rhoads DC, Germano JD, Solan M and others(2001) Sediment profile imagery as a benthic monitor-ing tool: introduction to a ‘long-term’ case history evalu-ation (Galway Bay, West Coast of Ireland). In: Aller JY,Woodin SA, Aller RC (eds) Organism−sediment inter -actions. The Belle W. Baruch Library in Marine Science,Vol 21. University of South Carolina Press, Columbia,SC, p 43−62

Kristensen E (2000) Organic matter diagenesis at theoxic/anoxic interface in coastal marine sediments, withemphasis on the role of burrowing animals. Hydrobio -logia 426: 1−24

Kurtz JC, Jackson LE, Fisher WS (2001) Strategies for evalu-ating indicators based on guidelines from the Environ-mental Protection Agency’s Office of Research andDevelopment. Ecol Indic 1: 49−60

Lefebvre S, Harma C, Blin JL (2009) Trophic typology ofcoastal ecosystems based on δ13C and δ15N ratios in anopportunistic suspension feeder. Mar Ecol Prog Ser 390: 27−37

Marinov D, Galbiati L, Giordani G, Viaroli P, Norro A, Ben-civelli S, Zaldívar JM (2007) An integrated modellingapproach for the management of clam farming in coastallagoons. Aquaculture 269: 306−320

McKindsey CW, Thetmeyer H, Landry T, Silvert W (2006)Review of recent carrying capacity models for bivalveculture and recommendations for research and manage-ment. Aquaculture 261: 451−462

McKindsey CW, Archambault P, Callier MD, Olivier F (2011)Influence of suspended and off-bottom mussel culture onthe sea bottom and benthic habitats: a review. Can J Zool89: 622−646

Mulsow S, Krieger Y, Kennedy R (2006) Sediment profileimaging (SPI) and micro-electrode technologies inimpact assessment studies: example from two fjords inSouthern Chile used for fish farming. J Mar Syst 62: 152−163

Muxika I, Borja A, Bonne W (2005) The suitability of themarine biotic index (AMBI) to new impact sources alongEuropean coasts. Ecol Indic 5: 19−31

Newell RIE (2004) Ecosystem influences of natural and cultivated populations of suspension-feeding bivalvemolluscs: a review. J Shellfish Res 23: 51−61

Nilsson HC, Rosenberg R (2000) Succession in marine ben-thic habitats and fauna in response to oxygen deficiency: analysed by sediment profile-imaging and by grab samples. Mar Ecol Prog Ser 197: 139−149

Nizzoli D, Welsh DT, Fano EA, Viaroli P (2006) Impact ofclam and mussel farming on benthic metabolism andnitrogen cycling, with emphasis on nitrate reductionpathways. Mar Ecol Prog Ser 315:151–165

Nobre AM, Ferreira JG, Newton A, Simas T, Icely JD, NevesR (2005) Management of coastal eutrophication: integra-tion of field data, ecosystem-scale simulations andscreening models. J Mar Syst 56: 375−390

Nobre AM, Musango JK, de Wit MP, Ferreira JG (2009) Adynamic ecological−economic modeling approach foraquaculture management. Ecol Econ 68: 3007−3017

OECD (Organisation for Economic Co-operation and De -velopment) (2000) Framework to measure sustainabledevelopment. OECD, Paris

Officer CB, Sayda TJ, Mann R (1982) Benthic filter feeding: a natural eutrophication control. Mar Ecol Prog Ser 9: 203−210

Olsen SB (2003) Frameworks and indicators for assessingprogress in integrated coastal management initiatives.Ocean Coast Manag 46: 347−361

Pearson TH, Rosenberg R (1978) Macrobenthic successionin relation to organic enrichment and pollution of themarine environment. Oceanogr Mar Biol Annu Rev 16: 229−311

Plew DR (2011) Shellfish farm-induced changes to tidal cir-culation in an embayment, and implications for sestondepletion. Aquacult Environ Interact 1: 201−214

Prins TC, Smaal AC, Dame RF (1998) A review of the feedbacks between bivalve grazing and ecosystem pro-cesses. Aquat Ecol 31: 349−359

Read PA, Fernandes TF, Miller KL, Davies IM, Rodger GK(2000) MARAQUA—the monitoring and regulation ofmarine aquaculture in Europe. J Appl Ichthyol 16:137–229

Rice J (2003) Environmental health indicators. Ocean CoastManag 46: 235−259

Rice J, Trujill V, Jennings S, Hylland K, Hagstrom O,Astudillo A, Jensen JN (2005) Guidance on the applica-tion of the ecosystem approach to management of humanactivities in the European marine environment. ICESCoop Res Rep 273. International Council for the Explo-ration of the Sea, Copenhagen

Rochet MJ, Trenkel VM (2003) Which community indicatorscan measure the impact of fishing? A review and pro -posals. Can J Fish Aquat Sci 60: 86−99

Rosland R, Bacher C, Strand Ø, Aure J, Strohmeier T (2011)Modelling growth variability in longline mussel farms asa function of stocking density and farm design. J Sea Res66: 318−330

Rudd MA (2004) An institutional framework for designingand monitoring ecosystem-based fisheries managementpolicy experiments. Ecol Econ 48: 109−124

Segnestam L (2002) Indicators of environment and sustain-able development. The World Bank Environment De -partment, Washington, DC

Shumway SE (ed) (2011) Aquaculture and the environment.Wiley-Blackwell, Chichester

Smaal AC, Prins TC (1993) The uptake of organic matter andthe release of inorganic nutrients by bivalve suspensionfeeder beds. In: Dame RF (ed) Bivalve filter feedersin estuarine and coastal ecosystem processes. SpringerVerlag, Berlin, p 271−298

Soto D, Aguilar-Manjarrez J, Brugère C, Angel D and others(2008) Applying an ecosystem-based approach to aqua-culture: principles, scales and some management mea-sures. In: Soto D, Aguilar-Manjarrez J, Hishamunda N(eds) Building an ecosystem approach to aquaculture.FAO/Universitat de les Illes Balears Expert Workshop,7−11 May 2007, Palma de Mallorca, Spain. FAO FishAquacult Proc 14. FAO, Rome, p 15−35

212


Souchu P, Vaquer A, Collos Y, Landrein S, Deslous-PaoliJM, Bibent B (2001) Influence of shellfish farming activi-ties on the biogeochemical composition of the water col-umn in Thau lagoon. Mar Ecol Prog Ser 218: 141−152

Spangenberg JH (2002) Environmental space and the prismof sustainability: frameworks for indicators measuringsustainable development. Ecol Indic 2: 295−309

Stevens CL, Petersen JK (2011) Turbulent, stratified flowthrough a suspended shellfish canopy: implications formussel farm design. Aquacult Environ Interact 2: 87−104

Stigebrandt A (2011) Carrying capacity: general principlesof model construction. Aquacult Res 42: 41−50

Strohmeier T, Aure J, Duinker A, Castberg T, Svardal A,Strand Ø (2005) Flow reduction, seston depletion, meatcontent and distribution of diarrhetic shellfish toxins in along-line mussel (Mytilus edulis) farm. J Shellfish Res 24: 15−23

Warwick RM, Clarke KR (1993) Increased variability as a

symptom of stress in marine communities. J Exp Mar BiolEcol 172: 215−226

Weise AM, Cromey CJ, McKindsey CW, Callier MD,Archambault P (2009) Shellfish-DEPOMOD: modellingthe biodeposition from suspended shellfish aquacultureand assessing benthic effects. Aquaculture 288: 239−253

Whitmarsh D, Wattage P (2006) Public attitudes towards theenvironmental impact of salmon aquaculture in Scot-land. Eur Environ 16: 108−121

WWF (World Wildlife Fund) (2007) Benchmarking studyon International Aquaculture Certification Programmes.WWF, Zurich and Oslo

WWF (2010) Bivalve Aquaculture Dialogue standards.Available at www.worldwildlife.org/what/globalmarkets/aquaculture/ WWFBinaryitem17872.pdf

Zurek MB, Henrichs T (2007) Linking scenarios across geo-graphical scales in international environmental assess-ments. Technol Forecast Soc Change 74: 1282−1295

213

Editorial responsibility: Ioannis Karakassis, Heraklion, Greece

Submitted: September 5, 2011; Accepted: April 4, 2012Proofs received from author(s): May 1, 2012

an ecosystem-based approach and management framework for

Documents