monitoring of coastal and transitional waters under the e ... · water, and coastal and...

Monitoring of coastal and transitional watersunder the E.U. Water Framework Directive

J. G. Ferreira & C. Vale & C. V. Soares & F. Salas &

P. E. Stacey & S. B. Bricker & M. C. Silva &

J. C. Marques

Received: 1 January 2006 /Accepted: 22 January 2007# Springer Science + Business Media B.V. 2007

Abstract A set of guidelines are presented for thedefinition of monitoring plans in coastal and transi-tional (estuarine and lagoonal) systems subject to theEuropean Union Water Framework Directive – WFD(2000/60/EC). General principles of best practice inmonitoring are outlined, including (a) the definition ofthree types of broad management objectives: waterquality, conservation, and human use, to which thegeneral public may easily relate. These will define thecore and research indicators (WFD quality elements)to be used for monitoring; (b) priorities and optimi-sation in a (financially and logistically) resource-constrained environment; (c) quality assurance; and(d) assessment of monitoring success: this shouldfocus on the outputs, i.e. the internal audit of themonitoring activity, and on the outcomes. The latter

component assesses programme effectiveness, i.e.environmental success based on a set of clearly-defined targets, and informs management action. Thesecond part of this work discusses the approach andactions to be carried out for implementing WFDsurveillance, operational and investigative monitor-ing. Appropriate spatial and temporal scales forsurveillance monitoring of different indicators aresuggested, and operational monitoring is classifiedinto either screening or verification procedures, withan emphasis on the relationship between drivers,pressure, state and response. WFD investigativemonitoring is interpreted as applied research, andthus guidelines cannot be prescriptive, except insofaras to provide examples of currently acceptableapproaches. Specific case studies are presented for

Environ Monit AssessDOI 10.1007/s10661-007-9643-0

J. G. Ferreira (*)IMAR – Institute of Marine Research, IMAR-DCEA,Fac. Ciências e Tecnologia, Qta Torre, 2829-516Monte de Caparica, Portugale-mail: [email protected]

C. ValeIPIMAR – Instituto de Investigação das Pescas e do Mar,Av. de Brasília, 1449-006Lisboa, Portugal

C. V. SoaresInstituto Hidrográfico, Rua das Trinas 49,1249-093 Lisboa, Portugal

F. SalasDepartamento de Ecología e Hidrología,Facultad de Biología, Campus Universitario de Espinardo,30100 Murcia, Spain

P. E. StaceyConnecticut Department of Environmental Protection,Bureau of Water Management, Hartford, CT, USA

S. B. BrickerNOAA – National Ocean Service, National Centers forCoastal Ocean Science, 1305 East West Highway,Silver Spring, MD 20910, USA

M. C. SilvaLNEC – Laboratório de Engenharia Civil,Av. do Brasil 101, 1700-066Lisboa, Portugal

J. C. MarquesIMAR – Institute of Marine Research,Faculdade de Ciências e Tecnologia,Universidade de Coimbra,3004-517 Coimbra, Portugal

both operational (coastal eutrophication control) andinvestigative monitoring (harmful algal blooms), inorder to illustrate the practical application of thesemonitoring guidelines. Further information is avail-able at http://www.monae.org/.

Keywords Coastal waters . Estuary . Transitionalwaters . Surveillance monitoring . Operationalmonitoring . Investigative monitoring . E.U.WaterFramework Directive . 2000/60/EC . U.S. CleanWaterAct .Management . MONAE

Introduction

The approval by the European Union of Directive2000/60/EC (European Community 2000), commonlyknown as the Water Framework Directive (WFD),established a comprehensive set of objectives forwater quality in European waters. This directiveestablishes a framework for community action inwater policy and management concerns, and appliesto all waters, including groundwater, inland surfacewater, and coastal and transitional waters. The variousobjectives defined in the WFD may be summarisedinto an overall goal: To achieve a good water statusfor all European Union waters by the year 2015.

The WFD stipulates that E.U. Member States mustestablish monitoring plans in order to assess theEcological Status of their water bodies. This is donethrough the assessment of a range of BiologicalQuality Elements (BQE) and Supporting Quality

Elements (SQE), that together lead to a classificationinto one of five Ecological Status classes, rangingfrom High Status to Bad Status. The WFD innovatesin a number of ways, amongst which is the require-ment of a definition of type-specific referenceconditions (Vincent et al. 2003), in recognition thatdifferent water types (e.g. different types of estuariesor lagoons) may be characterised by distinct defi-nitions of quality, with respect to environmentalmetrics such as phytoplankton biodiversity (e.g.Ferreira et al. 2005), benthic species composition(e.g. Borja et al. 2000, 2004a; Salas et al. 2004) andSQE (Bald et al. 2005). The complexity of the WFDrequired the production of a set of guidance docu-ments (European Commission 2006) and the scientificinterpretation of many of its requirements has resultedin publications addressing issues such as watermanagement (Mostert 2003), definition of waterbodies (Ferreira et al. 2006), or risk assessment forquality status targets (Borja et al. 2006).

Monitoring programmes will determine the com-pliance of E.U. Member States with the referenceconditions defined for each water type. Three types ofmonitoring programmes are defined in the WFD, eachaddressing different questions; these programmesconsequently vary in scope in both (1) time andspace, and (2) in the range of quality elements whichneed to be monitored (Table 1).

The goals and requirements of the U.S. CleanWater Act (CWA) are similar to those of the E.U.WFD, requiring states to regularly monitor and reporton the condition of coastal water bodies within theirjurisdiction (e.g. Gibson et al. 2000). The U.S.

Table 1 Types of monitoring defined in the WFD

Monitoring type Objectives

Surveillancemonitoring

Supplement and validate the assessment of the likelihood that transitional or coastal waters are failing to meetthe environmental quality objectives

Efficient and effective design of future monitoring programmesAssess long-term changes in natural conditions in order to distinguish between non-natural and naturalalterations in the ecosystem

Assess long-term changes resulting from widespread anthropogenic activityOperationalmonitoring

Establish the status of those bodies identified as being at risk of failing to meet their environmental objectivesAssess any changes in the status of such bodies resulting from the programmes of measures

Investigativemonitoring

Where the reason for any exceedences of environmental objectives is unknownWhere surveillance monitoring indicates that the objectives set under Article 4 for a body of water are not likelyto be achieved and operational monitoring has not already been established, in order to ascertain the causes ofa water body or water bodies failing to achieve the environmental objectives

To ascertain the magnitude and impacts of accidental pollution.

Environ Monit Assess

Environmental Protection Agency (USEPA) definesthe monitoring elements, which are detailed in EPA-approved state monitoring strategies. The monitoringprograms are implemented through a federal-statepartnership (USEPA 2003a). Monitoring programs aredesigned to assess attainment of state water qualitystandards and criteria, and are delegated to state watermanagement authorities. States report to EPA onwater quality status every two years as required bysection 305(b) of the CWA (USEPA 2003a). Thefocus is on water bodies that do not meet specificchemical, physical or biological standards and criteriaand are considered impaired. Identification of impair-ments, similar to designation of poor or bad statusunder the WFD, initiates management actions (re-sponse), including development of a managementplan, to bring the water body into compliance. Forchemical impairments, the water body is placed on alist identifying it as impaired under Section 303(d) ofthe CWA, which mandates listing of chemically-impaired water bodies, (USEPA 2003a). Section 303(d) requires a detailed investigation of the condition,including monitoring and development of a TotalMaximum Daily Load (TMDL) that establishesthresholds for chemical pollutant loading to a waterbody and a plan to attain that load (USEPA 2003b). Inmany ways, the U.S. CWA framework parallels theE.U. WFD requirements with the bi-annual monitor-ing and reporting equivalent to the surveillance moni-toring of the WFD and the more intensive monitoringand TMDL development equivalent to the WFDoperational and investigative monitoring. Althoughnot as specifically constrained by Section 303(d)requirements, similar activities are undertaken at thestate level to address physical and biological impair-ments when a relationship to a specific chemicalpollutant does not exist or is uncertain.

Monitoring of coastal systems has been executedfor many years within E.U. Member States prior tothe approval of the WFD (e.g. De Jong et al. 1999; DeJonge et al. 2006; MEMG 2004). Work on monitoringapproaches for surface waters under the WFD isongoing, for instance in the application of GIS (Chenet al. 2004); In freshwaters, Friberg et al. (2005) havepublished a comprehensive plan for quality monitor-ing in Danish streams. The Monitoring Plan ForPortuguese Coastal Waters – Water Quality andEcology (MONAE) project (Ferreira et al. 2005)was commissioned by the government of Portugal to

prepare a blueprint for WFD-compliant monitoring ofcoastal and transitional systems. This paper presentssome of the findings from MONAE, and addressesthe following objectives:

1. Definition and discussion of aspects which are ofgeneral application to any monitoring plan forsurface waters;

2. Review and interpretation of the three differenttypes of monitoring required by the WFD forcoastal and transitional waters, bearing in mindoptimisation issues due to logistic and financialconstraints;

3. Proposal and justification of appropriate temporaland spatial coverage for different BQE/SQE andillustration of this through the application of casestudies.

Monitoring guidelines

The following topics are covered as a generalframework for any monitoring activity:

1. Definition of appropriate objectives2. Setting priorities and optimisation3. Implementation of quality control4. Assessment of monitoring success

Definition of appropriate objectives

Although the general objective of monitoring speci-fied in the WFD is to verify compliance with waterquality objectives, or to establish the reasons for non-compliance so that appropriate measures may be putin place where applicable, a monitoring plan shouldexamine these questions in broader terms, from thestandpoint of ecosystem health. Monitoring activitiesthat address a broad set of aims use indicators asproxies for these. In the WFD, these indicators mustinclude the appropriate BQE and SQE, and mayinclude others.

The indicators shown in Fig. 1 may have differentlevels of aggregation, ranging from, for example,combined indices of eutrophication or benthic qualitystatus to the concentration of a particular parametersuch as dissolved oxygen, and may be definedcollectively as Environmental Quality Proxies (EQP).In theWFD these correspond to different combinationsof BQE and SQE. Relevant objectives should be


defined for management of Transitional and CoastalWaters (TCW), forming a set of goals which may needto be harmonised in time, space, and within theallowable EQP thresholds.

Three broad groups of management objectives maybe defined:

1. Water quality objectives – e.g. (a) Restore andmaintain a productive ecosystem with no adverseeffects due to pollution; (b) Minimize health risksassociated with contact water uses; (c) Estimateadverse impacts of eutrophication, includinghypoxia resulting from human activities;

2. Conservation objectives – e.g. (a) Maintain on alandscape level the natural environment of thewatershed; (b) Protect existing habitat categorieswithin the watershed to preserve and improveregional biodiversity;

3. Human use objectives – e.g. (a) Support water-related recreation while preserving the economicviability of commercial endeavours; (b) Encour-age sustainable lifestyles within the watershed,whereby human uses are balanced with ecosystem

protection; (c) Empower citizens in the protectionand stewardship of the coastal system (estuary,lagoon etc) and its watershed;

General objectives such as these have broad appeal,are easy to explain to a wide audience, and should beconsidered as bridges between ecosystem managementat a technical and scientific level, political decision-makers and the public at large. There is therefore arequirement that monitoring plans address these broadobjectives, using EQP as assessment tools.

Core and research biological and supporting qualityelements

A number of monitoring plans for coastal systems inthe E.U. and U.S. have identified several types ofindicators that can be used, which may be applied in acomplementary manner to address the issues underconsideration. These are typically divided into coreand research indicators, and are evaluated in distincttypes of monitoring plans. This fits in well with theconcepts outlined in the WFD and developed in

Aimse.g. Establish quality status ofwaterbody

Indicatorse.g. Phytoplankton composition, biomass and abundance

Activitiese.g. Technical sampling program for Public participation in visual chlorophyll a and species analysis or other types of sampling

Government, public• Water Quality• Conservation• Human Use

Managers• Biological Quality Elements• Supporting Quality Elements

Indicese.g. ASSETS, AMBI, others recommended in TICOR etc

Technicians• Biological Parameters• Supporting Chemical Parameters• Hydromorphology• Habitat• …

Targets

Aggregation

Detail

Objectives

Aimse.g. Establish quality status ofwaterbody

Indicatorse.g. Phytoplankton composition, biomass and abundance

Activitiese.g. Technical sampling program for Public participation in visual chlorophyll a and species analysis or other types of sampling

Government, public• Water Quality• Conservation• Human Use

Managers• Biological Quality Elements• Supporting Quality Elements

Indicese.g. ASSETS, AMBI, others recommended in TICOR etc

Technicians• Biological Parameters• Supporting Chemical Parameters• Hydromorphology• Habitat• …

Targets

Aggregation

Detail

ObjectivesFig. 1 Conceptual relation-ship between aims, indices,indicators and activities.ASSETS: ASSessment ofEstuarine Trophic Status(Bricker et al. 2003); AMBI:Marine Biotic Index (Borjaet al. 2000); TICOR: Typol-ogy and Reference Condi-tions for Portuguese CoastalSystems (Bettencourt et al.2003)


various guidance documents, i.e. that for surveillancemonitoring the full spectrum of BQE/SQE1 needs tobe covered, for operational monitoring the indicatorsneed to be far more targeted, and in the case ofinvestigative monitoring the focus is on the detailedunderstanding of a specific issue.

An example of the types of indicators used in theBarnegat Bay (New Jersey, U.S.) monitoring plan isshown in Table 2 (Barnegat Bay National EstuaryProgram 2002). The distinction between primary(high-profile) and secondary (internal-use) is similarto the higher and lower levels of aggregationillustrated in Fig. 1. The use of indicators to definepressure, state and response characteristics and trendshas grown in popularity. Indicators, particularlybiological indicators that are more charismatic thanchemical concentration data, for example, can providemore of an ecosystem perspective of conditions inestuarine and coastal waters that scientists, managers,politicians and the public find relevant and useful.

The Barnegat Bay example identifies a mix ofindicators that relate to this wide constituency ofscientists (e.g., chemical pollution and biologicaleffects), managers (e.g., pollutant sources and landcover changes), and politicians and the public (e.g.,fish and shellfish abundance and value, beachclosures, and toxic contaminants in seafood). The fullpicture of a coastal system incorporates all theseindicators to define cause and effect relationships thatlead to necessary management outcomes.

Furthermore, TCW monitoring programmes oftenneed only slight modifications to ensure that a broadsuite of useful indicators is built from the underlyingparameter and media monitoring, that meets WFDobjectives for comprehensive monitoring to assessecosystem health status.

Priorities and optimisation

Monitoring plans must be established for a compre-hensive coverage of transitional and coastal waterbodies. The monitoring activities to be carried outconstitute a serious additional workload on thetechnical and scientific communities in E.U. MemberStates, and due to logistic and/or financial constraintsit may be necessary to prioritise different monitoringactivities according to the management issues at hand.Additionally, models and prior monitoring efforts mayprovide enough insight into an ecosystem to improveefficiency by reducing sampling in time and space,using a more minimalist approach but still achievingmonitoring objectives.

Figure 2 shows a decision-tree that may be used todefine guidelines for prioritising monitoring activities.This approach takes into consideration:

(a) The definitions contained in the WFD forselection criteria of monitoring types – thesedefinitions are sometimes ambiguous;

(b) The pressure (anthropogenic or non-anthropo-genic) on the system;

(c) The susceptibility of the system, dependent onfactors such as freshwater flushing time and tidalmixing;

1As required for the type-specific definition of referenceconditions. Some elements may be excluded, e.g. due to highnatural variability.

Table 2 Indicator list (abridged) for the Barnegat Bay monitoring plan

Primary indicators (high-profile indicators) Secondary indicators (internal-use indicators)

Submerged aquatic vegetation distribution, abundance,and health

Temperature and salinity

Land use/land cover change Dissolved oxygen and nutrientsSignature species TurbidityWatershed integrity Phytoplankton abundance and composition & chlorophyll

a concentrationsShellfish beds Macrophyte abundanceBathing beaches Shellfish and finfish abundanceWater-supply wells/drinking water Benthic community structureHAB Toxic contaminants in aquatic biota and sedimentsFreshwater inputs Rare plant and animal populations


(d) The state of the system, assessed by means ofEnvironmental Quality Proxies, i.e. BQE andSQE.

The monitoring actions, whilst important to definingpressure and state conditions, are the Responsecomponent of this framework, and correspond todifferent types of monitoring. In Fig. 2 these arediscriminated by monitoring type, and colour-codedaccording to priority. Surveillance monitoring is notsubject to ranking according to this scheme, since it isa requirement of WFD river basin plans.

Implementation of quality control

Data quality is an important consideration for anymonitoring programme to ensure objectives are met

and conclusions are not misled by inaccurate data.The USEPA provides detailed guidance for QualityAssurance Project Plans (QAPP) that cover all aspectsof programme structure, quality assurance and con-trol, and data analysis and reporting (USEPA 2000,2001a). USEPA promotes a system that brings a finalproject design through policy, organisational, pro-grammatic and, finally, project implementation com-ponents. The guidance documents include anevaluation of quality control in environmental mod-elling, a critical component of monitoring andassessment (USEPA 2002).

QAPPs need to consider all aspects of themonitoring activity, emphasizing standard and recog-nizable elements from planning through implementa-tion and final assessment. As such, USEPA approvalrequires compliance with the following checklist:

Fig. 2 Decision-tree for se-lection of different types ofmonitoring programmes

6Sensu e.g. Bricker et al.(2003)


❑ The project technical and quality objectives areidentified and agreed upon;

❑ The intended measurements, data generation,or data acquisition methods are appropriate forachieving project objectives;

❑ Assessment procedures are sufficient forconfirming that data of the type and qualityneeded and expected are obtained;

❑ Any limitations on the use of the data can beidentified and documented.

Within this framework, QAPPs need to define andjustify an appropriate project management structure,sound data generation and acquisition methodologies,a reasonable and statistically validated assessment andoversight procedure for quality assurance and control,and a process for ensuring data are valid and usablefor the stated purposes and objectives of theprogramme. Of particular relevance is sound experi-mental design. The traditional components of aprogramme answer questions about the type andnumbers of samples, the design of the network,locations, frequencies of collection, media sampled,and the parameters included. Standard proceduresneed to be specified – sample methods, handling andcustody, and analytical methods.

Adequate quality control and assurance measuresspecific to the programme design that quantifyprecision, accuracy, bias, detection limits, proceduralerror, etc. must be included, together with plans torespond to any problems that arise. Typically, analyt-ical programmes rely on duplicates, splits, blanks,spikes, and reference samples, among others, duringboth field and laboratory operations. Review isfollowed from field and laboratory procedures intodata analysis and verification.

Appropriately validated banking of raw data is afundamental component of the quality assuranceprocess. Data collected in a monitoring programmemust be stored in such a way as to allow a variety oftreatments to be carried out. This includes, but is notlimited to, statistical analyses, use in GIS and modelcalibration and validation. Finally, quality assurancefor environmental models is an essential componentof the complete monitoring process cycle. Thisbecomes particularly significant as simulation resultsbecome progressively more integrated in regulatory

activities. Key points highlighted by the EPA guid-ance and other sources include:

(a) Suitability for purpose(b) Internal consistency(c) Adequate calibration, validation and sensitivity

testing(d) Appropriate documentation(e) Ease of use, including data input and output

handling

Assessment of monitoring success

The success of each monitoring plan must be assessedin a clear way, providing a mechanism for evaluatingthe cost-benefit of the monitoring activity and formaking necessary adaptations or corrections for futureimprovement.

Each monitoring plan must set out a number ofobjectives, which may be grouped into two differenttypes:

1. Outputs: These are verifiable targets, which maybe related to the terms of reference (i.e. Are thegoals and objectives of the plan being met?), andis effectively an internal audit – verification thatwould include compliance with the various termsof reference for time, space, parameters, method-ology, etc. This answers programmatic questionssuch as: (a) Is the sampling covering the estuaries/coastal systems specified in the plan? (b) Is thestrategy defined for a particular system (e.g.sampling according to a salinity gradient, partic-ular vertical profiles or seasonality being fol-lowed? (c) Are the parameters being measured asrequired by theWFD? (d) Are methodology issues(intercalibration of methods, etc.) being handledas recommended?

2. Outcomes: This component assesses programmeeffectiveness, i.e. environmental success andinforms management action. A distinct set oftargets, based around specific ecological qualityachievements, must answer questions such as: (a)Are shellfish/finfish areas increasing/decreasing?(b) Are salt marsh areas increasing/decreasing?(c) How is the frequency/spatial scope of elevatedchlorophyll a evolving? (d) What are the observ-able trends for harmful algal bloom (HAB)


events? (e) Are elevated nutrients correlated withelevated chlorophyll a? These questions should becentered around the BQE/SQE, and the indicesinto which these are aggregated.

As an example, for assessment of chlorophyll a (BQE)and dissolved oxygen (SQE), which are respectivelyprimary and secondary symptoms of eutrophication,monitoring success might be evaluated (1) at the outputslevel by examination of compliance with monthlysampling within water bodies covering three estuarinesalinity zones, considering appropriate depth profiling,analytical methods, etc.; and (2) at the outcomes level bydetermining whether the data collected provided suffi-cient information to answer questions on whether theimpacted areas and deviation from state at referenceconditions were increasing, and whether a correlationwith nutrient loading could be established.

Figure 3 illustrates an (abridged) form taken fromthe Tillamook Bay (Oregon, U.S.) monitoring plan(Tillamook Bay National Estuary Program 1999).This plan provides a good example of a programmethat addresses many of the design and qualityassurance concerns identified above. The survey foreelgrass (Zostera sp.) in Tillamook Bay is developedon the basis of the metadata presented in this form,which include a clear definition of the metrics used toevaluate monitoring success.

The form states the general objective (outcome, asdefined above) of the survey, allowing hypotheses(here adapted and posed as null) such as “Thedistribution of eelgrass is unchanged over a historicaltime period” or “The abundance of eelgrass is notbeing affected by nutrient enrichment” to be tested bymanagers. It additionally includes a managementobjective, “No net decline” – whilst this is not strictlya monitoring consideration, it is very useful to includethe general management objective in such a list.

There are a number of logistic and administrativefields, and finally a sufficiently complete set of outputindicators to allow a clear definition of the monitoringactivity and subsequent internal audit.

Surveillance monitoring

Definition and objectives

The WFD states: “For each period to which a riverbasin management applies a surveillance monitoring

shall be established. The objective of the surveillancemonitoring is to provide information for:

(1) supplementing and validating the assessment ofthe likelihood that transitional or coastal waterswill fail to meet the environmental qualityobjectives;

(2) the efficient and effective design of futuremonitoring programmes;

(3) the assessment of long-term changes in naturalconditions in order to distinguish between non-natural and natural alterations in the ecosystem;

(4) the assessment of long-term changes resultingfrom widespread anthropogenic activity.

The results of surveillance monitoring shall bereviewed and used in combination with the impactassessment to determine or adjust requirements forcurrent and other monitoring programmes in the riverbasin management plans.

On the basis of these results, the risk of failing tomeet WFD environmental objectives shall be evalu-ated in the surveyed water bodies and an operationalmonitoring programme established. Before imple-menting operational programmes and to ascertain thecauses of a water body failing to achieve theenvironmental objectives, investigative monitoringshall be considered, which may provide insight intoreasons for any unknown excess.”

Design of a surveillance monitoring programme

The foremost concerns in the design of a surveillancemonitoring programme are that (1) transitional andcoastal water sampling stations within each riverbasin district are sufficient in number; and (2)observations are frequent enough to provide anassessment of the overall water status.

Surveillance monitoring shall be carried out foreach monitoring site for a period of one year duringthe six-year period covered by a river basin manage-ment plan, unless the previous surveillance monitor-ing exercise showed that the body concerned hasreached good status and there is no evidence ofchanges in impacts. In these cases, surveillancemonitoring shall be carried out once every three riverbasin management plans, i.e. every 18 years. Moni-toring shall include the quality elements listed inTable 3, and indicated in WFD Annex V.


Spatial and temporal domain

The frequency of observations used over the surveil-lance monitoring period shall be sufficient to obtain arepresentative picture of the water body status. Thenumber of observations at each station will dependupon the variability in parameters resulting from bothnatural and anthropogenic conditions. The under-standing of the time scales of processes relevant towater quality status, obtained from previous monitor-ing programmes or literature reviews, informs anappropriate choice of monitoring frequency. It isrecommended that the times at which monitoring isundertaken shall be selected in order to ensure that theresults reflect changes in the water body due toanthropogenic pressure rather than other influences.

The minimum monitoring frequencies indicatedin Annex V of the WFD may not be adequate orrealistic for TCW. There will generally be a lowerlevel of confidence in most transitional systemswhen compared to freshwater because of the muchhigher natural variability and heterogeneity, there-fore more samples may also be needed. Addition-ally, areas of special conservation interest, e.g.Natura2000 sites (Wätzold and Schwerdtner 2005;Weiers et al. 2004), may require a fuller samplingprogramme to verify compliance with complementarylegislation such as the 92/43/EEC (Habitats) Direc-tive. Where possible, a horizontal approach tomonitoring should be taken, allowing resource opti-misation in meeting the requirements of multipledirectives.

Fig. 3 Submerged Aquatic Vegetation survey, Tillamook Bay National Estuary Program (2005), abridged (Tillamook Bay NationalEstuary Program 1999)


Table 3 Proposed surveillance monitoring frequencies of quality elements in coastal and transitional waters

Water bodies(a) Open coastal water bodiesQuality elements No influence of freshwater Submarine

canyonInfluence of freshwater Influence of urban

outfallsBiologicalPhytoplankton Seasonal Every six

monthsSeasonal Seasonal

Other aquatic flora Annual if applicable Notapplicable

Annual if applicable Every six months ifapplicable

Macro invertebrates Annual Notapplicable

Every six months Every six months

Fish Not applicable Notapplicable

Not applicable Not applicable

HydromorphologyDepth variation 6 years 18 years 6 years 6 yearsStructure of the bed 6 years 18 years 6 years 6 yearsStructure of the intertidalzone

6 years Notapplicable

6 years 6 years

Tides Continuous Notapplicable

Continuous Continuous

Currents and flows 6 years 18 years 6 years 6 yearsWave exposure Continuous for one year

every six yearsNotapplicable

Continuous for one yearevery six years

Continuous for one yearevery six years

Physico-chemicalTransparency/Turbidity Seasonal Seasonal Seasonal SeasonalThermal conditions Seasonal Seasonal Seasonal SeasonalDissolved oxygen Seasonal Seasonal Seasonal SeasonalSalinity – Seasonal Seasonal SeasonalNutrient status Seasonal Seasonal Seasonal Seasonal

Special Pollutants2

Other pollutants Annual(2) Annual(2) Annual(2) Annual(2)

Priority substances Annual(2) Annual(2) Annual(2) Annual(2)

(b) Restricted coastal and transitional water bodiesQuality element FrequencyBiologicalPhytoplankton (biomassand abundance)

Monthly

Phytoplankton speciescomposition

Every six months

Other aquatic flora SeasonalMacro invertebrates Every six monthsFish3,4 Seasonal

Hydromorphology VariablePhysico-chemicalThermal conditions MonthlyDissolved oxygen MonthlySalinity MonthlyNutrients Monthly

Special Pollutants5

Other pollutants Every six monthsPriority substances Seasonal

2 Sampling should be carried out in tissues of fish and shellfish and in sediments3 In the WFD, applicable only to transitional waters4 Observations shall where possible be synoptic with monitoring programmes related to the sustainable exploitation of commercial fish species5 Sampling should be conducted in suspended particulate matter, sediments and tissues of fish and shellfish


Open coastal waters

Although the WFD applies only to a seaward limit ofone nautical mile from the baseline of territorialwaters, this may comprise a rather large area andvolume of coastal water, in particular for nationswhich use straight lines across promontories or capesas the baseline (e.g. France, Ireland, Germany,Portugal, U.S.A.). These more extensive coastalwaters are not directly influenced by river inputs orsewage discharges, and most of the changes inphysico-chemical and biological parameters are dueto natural conditions. Monitoring frequencies shall bechosen to achieve an acceptable level of long-termsurveillance.

Table 3 contains a set of of proposals developed forPortuguese TCW in the MONAE programme, takinginto account the Monitoring Guidance prepared by theEuropean Commission (2003). The rationale is ingeneral terms to optimise the information obtained,whilst working within resource limitations. Thefrequencies suggested are taken to be minimalrequirements for adequately resolving the mainprocesses of interest, and for the successful applica-tion of water quality and ecological models.

Table 3a summarises a set of proposed samplingfrequencies of quality elements. The frequenciesdiffer, but sampling should always take place synop-tically, e.g. samples collected at three-month periodsfor a particular element should coincide with monthlysamples for relevant elements. The vertical samplingresolution should be determined according to thewater temperature and salinity gradients, but alwaysinclude at least a surface and a deep water sample(above and below the pycnocline for a stratified watercolumn).

Transitional and inshore coastal waters

These include estuaries and coastal waters in theproximity of estuaries or lagoons, where water statusis influenced by the magnitude of discharges as wellas by their tidal and seasonal fluctuations. Monitoringfrequencies of pelagic BQE and SQE shall take intoconsideration the tidal and seasonal variability. Ateach station in estuaries and coastal lagoons withpermanent connection to the sea it is recommendedthat all these parameters be measured at least at high

and low tide, supplemented by sampling at mid-ebband mid-flood where appropriate.

We propose that spatial resolution will be deter-mined on the basis of the water bodies defined foreach system (Ferreira et al. 2006) with at least onestation per water body. This addresses a “paradox”within the WFD, whereby “Member States must besure that all Water Bodies have Good EcologicalStatus but only a subset may be sampled.” Alterna-tively, monitoring may be used for testing thehypothesis: If Water Body X is at Good EcologicalStatus with certain pressures, Water Body Y is also atGood Ecological Status if it has (a) Similar suscep-tibility; (b) Equivalent pressure indicators; (c) Loadsin similar relative positions (e.g. with reference to thesalinity distribution).

The vertical resolution for BQE and SQE measuredin the water column should be determined (a) by thedepth of the station, and (b) by the degree ofstratification. The following general guidelines aresuggested for vertical sampling in transitional andinshore coastal waters, building on the generalorientation from the European Commission (2003):

❑ At stations with depth less than 2 m (withrespect to tidal datum), only mid-water sam-ples will be collected, unless there is clearsalinity and/or temperature stratification;

❑ At stations with depth of 2–4 m (with respectto tidal datum), surface and bottom sampleswill be collected. If clear salinity and/ortemperature stratification exists, an additionalmid-water sample will be taken;

❑ At stations with depth of 4–10 m (with respectto tidal datum), surface, mid-water and bottomsamples will be collected;

❑ At stations with depth greater than 10 m (withrespect to tidal datum), appropriate verticalprofiling will be used, based on salinity and/ortemperature stratification.

Table 3b summarizes a set of proposed samplingfrequencies for quality elements. The frequenciesshown for quality elements differ, but samplingshould always take place synoptically, e.g. samplescollected at three-month periods for a particularelement should coincide with monthly samples forrelevant elements.


Operational monitoring

Definition and objectives

Operational monitoring as defined by the WFDfocuses on two specific objectives: (a) Establish thestatus of those bodies identified as being at risk offailing to meet their environmental objectives; (b)Assess any changes in the status of such bodiesresulting from the programmes of measures. In bothcases, the objectives are to verify the status of a waterbody or set of water bodies, with respect to one ormore WFD quality elements.

Except in extreme cases of pressure across a rangeof substances (nutrients, metals, organic pollutants,etc.), this means that whereas surveillance monitoringis broader in scope, and as a rule less targeted,operational monitoring will generally focus on asubset of quality elements, e.g. primary and secondaryeutrophication symptoms in the case of nutrient-related problems (Bricker et al. 2003).

Identification of water bodies at risk and verificationof measures

The first objective (screening) of operational moni-toring is concerned with further investigation into awater body which is at risk of non-compliance withenvironmental objectives, i.e. which appears fromsurveillance monitoring data to be at moderate, pooror bad status for one or more quality elements.Operational monitoring is interpreted to be applicablemainly for water bodies diagnosed as being atmoderate status, where more detailed studies willhelp establish the status of the water body. Figure 2presents guidelines for managers to decide on theinception of operational monitoring as regards thefirst objective. It is intended as a primer for detaileddata analysis on a case by case basis, on which finaldecisions will be based.

The second objective (verification) is to verifypost-facto if management measures are working, i.e.from a Pressure-State-Response perspective, if areduction in pressure due to management responsehas resulted in the expected change in state. Theprediction of the change in state that will resultfrom changes in pressure may only be made usingthe same approaches used for definition of referenceconditions:

1. Comparison to historical data;2. Comparison to system(s) of identical type in

pristine condition;3. Application of ecological models;4. Heuristic evaluation.

The evaluation of the changes in status is madethrough the comparison of predictions andmeasurements.

Design of an operational monitoring programme

The guidelines for the design of operational monitor-ing will be determined by the quality elements that areunder scrutiny, and whether the monitoring is beingimplemented to address screening or verification. Thetwo objectives are discussed separately below, despitethe fact that there are some common points.

Operational monitoring for screening

The decision to implement operational monitoring forscreening purposes should be based on (a) the resultsof surveillance monitoring; (b) the pressures on a waterbody; or (c) both of these. Situations such as (1) highpressure combined with good state or (2) low pressurecombined with bad state (Fig. 2) clearly need furtherinterpretation. One of the key aspects in the design ofthis type of operational monitoring is the accurateassessment of anthropogenic pressure, including sourceapportionment, necessary in order to determine thepossible responses in various situations.

TCW exhibit changes in state that may appear to bedecoupled from the pressure on the system. Forinstance, in the case of coastal eutrophication:

1. The symptoms are diverse, variable in time andspace, may potentially be due to a range ofcauses, and vary greatly in severity;

2. Although there is an association between pressureand state, the relationship between them is stronglyinfluenced by geomorphology and hydrodynam-ics: TCW subject to similar nutrient-related pres-sure often exhibit totally different eutrophicationsymptoms, and in some cases no symptoms at all.Factors such as water residence time, tidal rangeand turbidity play a major role in determining thenature and magnitude of symptom expression;

3. Biological interactions, particularly due to graz-ing, may provide a top–down control of eutro-phication symptoms. These may occur in similar


types of TCW, due to natural variability, but alsodue to human activities such as shellfish aquacul-ture. In the latter case, selective filtration bybivalves may additionally affect biodiversity byaltering the phytoplankton species composition.

Conversely, Fig. 4 shows a situation where achlorophyll-rich water mass is moving inshore (left toright panes), illustrating how a potential HAB eventcauses impairment of coastal waters (Miller et al.2005). These events, caused for example by upwellingrelaxation, are observed regularly off the IberianPeninsula, Ireland, the U.S. NE and NW coasts, andmany other coastal areas. No reduction in pressure willcorrect phenomena of this type, since they are naturaloccurrences. Coastal waters subject to this type of“natural pressure” are only amenable to remedialmanagement measurements such as interdiction ofshellfish collection.

The design of a monitoring programme of this kind,which aims to screen water bodies and systems, musttherefore take into account (a) the measurement of state,where the design considerations are those indicated inthe surveillance monitoring section as regards particularquality elements; (b) the determination of pressure toestablish whether there is a match between pressure and

state; (c) source apportionment if required, in order toinform appropriate management measures.

Operational monitoring for verification

The design of an operational monitoring programmefor verification of compliance presupposes that thereis a clear hypothesis that relates the anthropogenicpressure to the ecological status.

In all cases, the null hypothesis being tested forone or more quality elements is: “The change inanthropogenic pressure as a result of managementresponse does not result in a change of state.” Thehypothesis is tested e.g. to verify whether decreasedpressure improves state, or if increased pressuredeteriorates state. In many cases, a reduction inpressure will result in an improvement of state, butin some cases, such as the HAB example in Fig. 4, itwill not. The key design consideration is therefore thetesting of this hypothesis, which must include anumber of steps below, following the operationalmonitoring programme:

1. Verify that the change in pressure is actuallyoccurring;

Fig. 4 Potential impairment of coastal waters by phytoplankton (represented as chlorophyll a) advected from offshore fronts (courtesyPlymouth Marine Laboratory)


2. Determine the state for the relevant qualityelement(s) using surveillance monitoring guide-lines;

3. Test the results against predictions of statechanges due to management response;

4. Accept or reject the effectiveness of measures.

Case study: Eutrophication control

An analysis of the potential effects of reduction innutrient loading for the Ria Formosa in southernPortugal is presented as a case study for implemen-tation of operational monitoring for verification, byusing an ecological model to simulate changes inpressure (due to measures) and testing the effective-ness on changes in system state.

The eutrophication status of the Ria Formosa wasdetermined by means of the ASSETS screening model(Bricker et al. 2003). The resulting eutrophicationgrade of Moderate, which corresponds to a WFD clas-sification of Moderate, was determined on the basis ofdata collected over a number of years for primary andsecondary eutrophication symptoms. In parallel, anecological model was developed for the Ria (Fig. 5), tosimulate water exchange, nutrient dynamics, pelagicand benthic production, and clam aquaculture, a majoruse of the system, for nine different compartments(boxes). The outputs of this model were used to drivethe screening model (Nobre et al. 2005). Fourscenarios were run on the research model: pristine,standard (simulates present loading), half and doublethe current nutrient loading.

The Ria Formosa has a short water residence time,and eutrophication symptoms are not apparent in thewater column. However, benthic symptoms areexpressed as excessive macroalgal growth and strongdissolved oxygen fluctuations in the tide pools. Thestandard simulation results showed an ASSETSgrade identical to the field data application. The ap-plication of the screening model to the other scenariooutputs showed the responsiveness of ASSETS tochanges in pressure, state and response, scoring agrade of High under pristine conditions, Good forhalf the standard scenario and Moderate for doublethe present loadings. The use of this hybrid approachallows managers to test the outcome of measuresagainst a set of well-defined metrics for the evalu-ation of state.

Figure 6 shows the results obtained for the researchmodel “green” scenario, corresponding to a 50%reduction in nitrogen loading. From an operationalmonitoring standpoint, the results generated by theresearch model could be compared to measured dataafter the implementation, using a variety of tech-niques, such as trend analysis or statistical compar-isons. More importantly, the ASSETS screeningmodel, which is a potentially valuable tool for theimplementation of the WFD in TCW (Borja et al.2006; Ferreira et al. 2005; Nobre et al. 2005), couldbe applied to the data set collected in the verificationprogramme, and the results compared with thescreening model classification shown in Fig. 6. Adetailed analysis of the application of these models isgiven in Nobre et al. (2005) and references therein.

Investigative monitoring

Cases where investigative monitoring is required

The Water Framework Directive specifies three caseswhere this type of monitoring is required:

1. Where the reason for any exceedences (ofEnvironmental Objectives) is unknown;

2. Where surveillance monitoring indicates that theobjectives set under Article 4 for a body of water arenot likely to be achieved and operational monitor-ing has not already been established, in order toascertain the causes of a water body or water bodiesfailing to achieve the environmental objectives;

3. To ascertain the magnitude and impacts ofaccidental pollution.

0 5 km

Depth (m):-5

0

> 40

1

23

45

67

8

9Box limitsInlets

0 5 5 kmkm

Depth (m):-5

0

> 40

1

23

45

67

8

9Box limitsInlets

Fig. 5 Ria Formosa, showing ecological model boxes (seeNobre et al. 2005)


The results of the monitoring would then be usedto establish a programme of measures to achieve theenvironmental objectives and specific measures nec-essary to remedy the effects of accidental pollution.Investigative monitoring will thus be designed for thespecific case or problem being investigated. In somecases it will be more intensive in terms of monitoringfrequencies and focused on particular water bodies orparts of water bodies, and on relevant qualityelements. Ecotoxicological monitoring and assess-ment methods would in some cases be appropriatefor investigative monitoring.

Investigative monitoring might also include alarmor early warning monitoring, for example, forprotection against accidental pollution. This type ofmonitoring could be considered as part of theprogrammes of measures required by Article 11.3(a)of the WFD and could include continuous or semi-

continuous measurements of a few chemical (such asdissolved oxygen) and/or biological (such as fish)determinants. Investigative monitoring may involveother determinants, sites and frequencies than surveil-lance or operational monitoring, as each programmewill be designed to assess a specific stress or impact.

Approaches in investigative monitoring

Investigative monitoring relies by definition on avariety of approaches, which will generally beconjugated to provide answers to the research ques-tions being asked (e.g. Allan et al. 2006; Dworaket al. 2005). Estuaries and coastal zones are oftenhighly energetic systems, due to the effects of riverdischarges, waves and tides, so the representation ofphysical processes is usually required to help clarifythe phenomena of interest. These physical tools

Index

Overall Eutrophic Condition (OEC)

ASSETS OEC: 3


ASSETS OEC: 3


ASSETS OEC:

Methods

PSM

SSM

PSM

SSM

PSM

SSM

Parameters Value Level of expression

Chlorophyll a 0.25 0.61Macroalgae 0.96 High

Dissolved Oxygen 0Submerged Aquatic 0.25 0.25Vegetation LowNuisance and Toxic 0Blooms



Chlorophyll a 0.25 0.38Macroalgae 0.50 Moderate


Index

MODERATE

MODERATE

MODERATELOW

4

Model

green

scenario

Index


ASSETS OEC: 3


ASSETS OEC: 3


ASSETS OEC:

Methods

PSM

SSM

PSM

SSM

PSM

SSM

Parameters Value Level of expression





Chlorophyll a 0.25 0.38Macroalgae 0.50 Moderate


Field data

Research

model

Index

MODERATE

MODERATE

MODERATELOW

40 % lower

4

Fig. 6 Application of ASSETS to various research model scenarios. PSM – Primary Symptoms Method; dSSM – SecondarySymptoms Method. These are combined in a matrix which provides a combined score using the area-weighted average for primaryeutrophication symptoms and the area-weighted maximum for secondary eutrophication symptoms, resulting in the classificationshown in the right column


should be used in conjunction with chemical andbiological techniques, which are selected according tothe objectives of the work.

Investigative monitoring of the marine environ-ment is by nature interdisciplinary, with the problemsaddressed being diverse, and constrained by differentlevels of understanding. Issues range e.g. from theinterpretation of the effects of an accidental oil spill,where most processes are well understood, to theunderstanding of changes in biodiversity, affectinge.g. phytoplankton or benthic species composition,which are rather poorly understood (Fig. 7). Thelevel of uncertainty in our understanding of underly-ing processes responsible for a particular environ-mental effect, and the corresponding apportioning ofhuman influence (which conditions the possibility andadequacy of the response), is thus a major factor inthe planning, execution and potential success of aninvestigative monitoring programme. In some casesthe same problem may have completely differentcauses, e.g. HAB may be clearly linked to anthro-pogenic influence (e.g. cyanophyte blooms in theBaltic Sea) and therefore amenable to managementmeasures (e.g. Turner et al. 1998), or perhaps beadvected to coastal areas from offshore fronts (e.g.Fraga et al. 1988; Joyce and Pitcher 2004), andtherefore not manageable in terms of emissioncontrol.

Overview of methodologies

Due to the constraints described above, it is onlyappropriate here to provide some examples ofmethodologies that may be used to address researchquestions (i.e. perform investigative monitoring) onbiological, supporting (chemical), and hydro-morphological quality elements. Additionally, itshould be recognised that (a) methodologies areconstantly under development (e.g. molecularprobes, chemotaxonomic methods, DNA barcodes,improved in situ instrumentation, remote sensing);and (b) future paradigm shifts will potentially makesome of these methods obsolete, as has occurred inthe past for example following the development ofremote sensing applications or mathematical models.It is therefore recommended that investigativemonitoring should always draw on the best availabletechniques, combining the state of the art in fielddeterminations, laboratory experiments and simula-tion models in order to provide the answers to theinvestigative monitoring questions posed by managersand scientists.

Application of biomarkers for investigativemonitoring

Biomarkers are discussed below as an example of apowerful tool for use in investigative monitoring,targeted at xenobiotics.

Biomarkers as an investigative monitoring tool

France and the U.K. have explored possibilities toinclude bioassays in the WFD; in most other countriesbioassays and biomarkers are applied at a researchlevel and/or in national monitoring programmesrelated to the OSPAR Commission for the Protectionof the Marine Environment of the North–East Atlantic(1999). Moreover, in the U.S. bioassays are a federalrequirement of state-delegated programmes for moni-toring point source effluents as part of the dischargelicensing process.

When warranted, field bioassays of water andsediment may be incorporated into state environmen-tal monitoring programmes to ascertain causes ofaquatic life use impairments and to track down toxiccontaminant sources. Sediment bioassays are also an

Understanding

Oil spill

Phytoplankton biodiversity

Hum

an in

fluen

ce

MoreLess

Mor

eLe

ss

Elevated nutrients

HABType 2

HABType 1

Harmful algal blooms (HAB)of various types (see text)

Understanding

Oil spill

Phytoplankton biodiversity

Hum

an in

fluen

ce

MoreLess MoreLess

Mor

eLe

ss

Elevated nutrients

HABType 2

HABType 1

Harmful algal blooms (HAB)of various types (see text)

Fig. 7 Examples of environmental problems in marine sys-tems, scaled by human influence and process understanding


integral component of testing materials to be dredgedif there is an expectation that the sediments are toxicbased on bulk analyses. In this sense, Borja et al.(2004b) point out the importance of the inclusion ofsediment analyses and the use of biomonitors in thedetermination of the quality standards of the WFD.

Although it is not specifically mentioned in theWFD, the WGBEC (2004) determined that there areclear opportunities for the use of bioassays in twoways:

1. To predict whether the chemical quality issufficient to achieve high ecological status, usingrisk analysis;

2. To determine whether chemicals are the cause ofnot achieving good ecological status.

Bioanalysis may be regarded as a partial replace-ment of chemical analysis of priority and/or otherrelevant substances, and prioritising locations forfurther chemical analysis. It is defined as theapplication of a small set of inexpensive and rapidassays representing various taxonomic groups and/ormodes of action applied to an extract of a water (orsediment) sample.

For bioassays, the WGBEC (2004) recommendedwhole sediment bioassays using the mud shrimpCorophium and the lugworm Arenicola marina, andbioassays of sediment pore waters, sea water elutri-ates, and sea water samples with bivalve embryos andthe planktonic copepod Acartia. These two types ofbioassays are non-contaminant specific and canprovide a retrospective interpretation of communitychanges. Moreover, WGBEC (2004) recommendeddifferent techniques for biological monitoring pro-grammes, some of which are pollutant specific (e.g.imposex for TBT, bulky DNA adduct formation andphytohemagglutinin bile metabolism methods forPAH and other synthetic organic compounds). Othermethods such as lysosomal stability, lysosomalretention and the “scope for growth” method, are notspecific and respond to a wide variety of contami-nants. The first two methods provide a link betweenexposure and pathological endpoints, and the “scopefor growth” method is a sensitive measure ofsublethal effects such as energy available for growthin bivalve molluscs (e.g. Savina and Pouvreau2004; Wang et al. 2005).

Sampling procedures and frequency

The sampling procedures and their frequency ofapplication will depend on the method selected toinvestigate the cause and the magnitude of specificstress or impact. In general terms, near-zone moni-toring (e.g., caging of bivalves for testing purposes)and wider area field surveys need to consider thebiology of the target organism. In particular, periodsof natural stress might need to be avoided, such asthe spawning period when there may be large fluxesof contaminants out of the organism with the releaseof eggs. The OSPAR Joint Assessment and Monitor-ing Programme (OSPAR Commission for the Protec-tion of the Marine Environment of the North–EastAtlantic 2004) has produced guidelines for generalbiological effects monitoring with technical annexesdescribing the methodology, sampling procedures andfrequency of different bioassays and biological meth-ods. Several European projects such as BEQUALM(2002) have developed protocols and procedures forbiological methods used in marine monitoring.

The role of target species

Bivalve molluscs have been one of the most frequent-ly used indicators to determine the existence andquantity of a toxic substance. The advantages of usingbivalves in environmental monitoring are: (1) widedistribution; (2) simplicity of sampling; (3) sedentarynature; (4) tolerance to a wide range of environmentalconditions; and (5) high potential for bioconcentrationof environmental toxicants due to high filtrationactivity.

Due to their sessile nature, wide geographicaldistribution and capability to detoxify when pollutionceases, mussels such as Mytilus, cockles such asCerastoderma and clams such as Donax have longbeen considered ideal for the detection of toxicsubstances in the environment. This broadly corre-sponds to the “Mussel Watch” concept (Goldberget al. 1978). Likewise, certain species of crustacea andpolychaete worms are considered capable of accumu-lating toxic substances.

Many fish species have also been used for thestudy of toxic pollution of the marine environment,due to their bioaccumulative capability and theexisting relationship between pathologies suffered by


benthic fish and the presence of polluting substances(e.g. USEPA 2001b; NOAA 2004). Commercial andrecreational fish species consumed by humans arealso indicators of potential human health risk whenspecified thresholds for contaminant accumulation areexceeded; the implications are readily understood bythe general public.

Case study: Harmful algal blooms

HAB are caused by many different species ofphytoplankton, and can have widely varying effects(Smayda 1997). They cause significant ecological andeconomic damage, for example through impacts onfisheries, aquaculture, human health and tourism.HABs may occur in open coastal waters and insemi-enclosed/enclosed systems, with a trend towardsthe occurrence of toxic algae in the former and highbiomass blooms in the latter. The investigation of thecauses and development of HAB events thus requiresa range of methodologies (Table 4).

The Ecology and Oceanography of Harmful AlgalBlooms ECOHAB (n.d.) programme (NOAA 2005a),initiated in the U.S. in 1995, included three broadresearch themes:

❑ Organisms – To determine physiological,biochemical and behavioural features influ-encing bloom dynamics;

❑ Environmental regulation – To determine andparameterise the factors controlling the onset,growth and maintenance of HABs;

❑ Food-web and community interactions – To de-termine the extent to which food webs and tro-phic structure affect and are affected by HABs.

A general aim of Ecology and Oceanography ofHarmful Algal Blooms ECOHAB (n.d.) was todevelop reliable models to predict bloom initiation,development, duration and toxicity. The MERHABprogram is a complementary initiative aimed atmonitoring and event response for HAB (NOAA2005b).The EUROHAB initiative is a similarprogramme that has been carried out in the E.U.since 1999, clustering research projects such asBIOHAB and ECOHARM.

As an example of the application of currentlyavailable investigative monitoring techniques, fieldand simulation studies in the Gulf of Maine, U.S.(McGillicuddy Jr et al. 2005), revealed a number ofphysical and biological mechanisms which play a key

Table 4 Examples of methodologies for investigative monitoring of HABs

Methodology Study objective

Physical oceanography (field studies, moorings) Vertical migration, water column stabilityRemote sensing Biomass and primary productivity determination, bloom tracking, model validationMicropaleontology Cyst distribution in sedimentsMolecular probes Toxicity assessment of facultative HAB speciesNumerical models Prediction of HAB population development and distributionCost-benefit analysis Evaluation of socio-economic costs of recurring HAB events

Table 5 Hypothesis-driven monitoring approach

Hypothesis Examples

H0: Changes in environmental status and health areunrelated to human pressures.

Red tide events in Portuguese coastal waters are unrelated to human use;Changes in estuarine turbidity are unrelated to increased organicproduction;

Given the nature of the Portuguese Atlantic coast, the only state changesin the coastal waters due to basin pressures are in xenobiotics in offshoresediments.

H0: Changes in (abiotic) Supporting Quality Elements arenot reflected in Biological Quality Elements.

State changes in dissolved nutrients do not have a discernible effect oneutrophication symptoms such as chlorophyll a concentration;

State changes of xenobiotics in offshore waters do not presently have adiscernible effect on marine biota.


role in the generation and maintenance of blooms ofthe toxic dinoflagellate Alexandrium fundyense:

1. Cysts of this species were found to germinate inbottom sediments far from shore;

2. Field mapping surveys revealed a large cystrepository situated offshore of Casco and Penob-scot Bays, at a depth of 150 m, with densitiesgreater than 20 times those in inshore waters;

3. The role of these deep-water cysts in coastalHABs was studied by means of a mathematicalmodel, which allowed the identification of anentrainment mechanism based partly on thebehaviour of the toxic organism and partly onthe wind-driven transport of a plume of lowsalinity water trapped in the surface layer;

4. The HAB cells germinated from deep-water cystsswim actively towards the light, enter the thinsurface layer and are advected to the coast due tofavourable onshore winds.

This case study illustrates the need to understandthe cause–effect relationships that underpin HABs,through a combination of research tools. Theaffected area would in all likelihood violate envi-ronmental objectives, but conventional measurescentered on the reduction of land-based nutrientdischarges would not be an appropriate managementresponse.

Conclusions

The implementation of the WFD raises many chal-lenges, which are widely shared by European UnionMember States. These include (1) the complexity ofthe text and the range of possible solutions toscientific, technical and practical questions; (2) theextremely demanding timetable; (3) the incompletetechnical and scientific basis with some fundamentalissues in Annex II and V, which need furtherelaboration in order to make the transition fromprinciples and general definitions to practical imple-mentation successful; and (4) a strict limitation ofhuman and financial resources. Most of these generalissues are also applicable to states of the U.S., underthe CWA.

A well-designed monitoring programme shouldendeavour to test one or more hypotheses, even if

the baseline objective is verification of the compli-ance status of a set of water bodies to the require-ments of the WFD. A statement of the hypotheses tobe tested, and the methodologies to be used toperform the tests must be a part of any monitoringplan. In the case of surveillance and operationalmonitoring, the hypotheses may address broad ques-tions, such as those listed in the left column ofTable 5. General hypotheses such as these may berefined to address specific issues, depending on thesystems under consideration. The right column givesexamples of this type of specification.

Operational monitoring will need to follow adrivers-pressure-state-response approach, particularlyas regards the verification of measures, whichconstitute the management response. This is consid-ered appropriate by other authors (e.g. Borja et al.2006; Elliott 2000) and is clearly an interface areabetween monitoring and ecological modelling. Inves-tigative monitoring programmes are by definitionaimed at hypothesis testing, in order to furtherunderstanding of key processes. These must thereforebe built on the basis of meaningful research questionsand take the form of scientific research projects.

The classification of monitoring by the WFD intothese three types is somewhat arbitrary, and inparticular it is questionable whether investigativemonitoring really fits into the general definition ofmonitoring or could be better described as appliedresearch. Nevertheless, the concepts and applicationof guidelines presented herein will help to structureand improve many monitoring schemes, avoiding afrequent situation where large volumes of data arecollected in a deus ex machina process, and subse-quently prove of little use to inform managementdecisions.

Partly this occurs because the various disciplinesrequired for an ecosystem approach to coastalmonitoring are not sufficiently integrated, even atthe level of diagnosis of state, and certainly as regardsinterfacing with land use, hydrological processes andeffluent discharge patterns in the relevant watersheds.The development of a more structured approach to themonitoring process will provide numerous directbenefits, and well-designed surveillance and opera-tional monitoring may additionally be leveraged forsubsequent complementary (and typically more spe-cific and detailed) investigative programmes.


Acknowledgements The authors are grateful to INAG forsupporting this work. We also wish to express our thanks to theMONAE project team for helpful comments, to PlymouthMarine Laboratory for access to remote sensing data, and totwo anonymous referees for their constructive comments on anearlier manuscript draft.

References

Allan, I. J., Vrana, B., Greenwood, R., Mills, G. A., Roig, B., &Gonzalez, C. (2006). A “toolbox” for biological andchemical monitoring requirements for the Europeanunion’s Water Framework Directive. Talanta, 69, 302–322.

Bald, J., Borja, A., Muxika, I., Franco, J., & Valencia, V.(2005). Assessing reference conditions and physico-chem-ical status according to the European Water FrameworkDirective: A case-study from the Basque Country (North-ern Spain). Marine Pollution Bulletin, 50, 1508–1522.

Barnegat Bay National Estuary Program (2002). Final compre-hensive conservation and management plan, completedunder a cooperative agreement between U.S. Environmen-tal Protection Agency Region 2 and the New JerseyDepartment of Environmental Protection, Barnegat BayNational Estuary Program, Ocean County College, TomsRiver, NJ. http://www.bbep.org/ccmp.htm.

Bettencourt, A., Bricker, S. B., Ferreira, J. G., Franco, A.,Marques, J. C., Melo, J. J., et al. (2003). Typology andreference conditions for Portuguese transitional andcoastal Waters. Development of guidelines for the appli-cation of the European Union Water Framework Directive.INAG/IMAR, 2003. http://www.ecowin.org/ticor.

Biological Effects Quality Assurance in Monitoring Pro-grammes (BEQUALM) (2002). http://www.bequalm.org/.

Borja, A., Franco, J., & Pérez, V. (2000). A marine biotic indexto establish the ecological quality of soft-bottom benthoswithin European estuarine and coastal environments.Marine Pollution Bulletin, 40(12), 1100–1114.

Borja, A., Franco, J., Valencia, V., Bald, J., Muxika, I.,Belzunce, M. J., et al. (2004a). Implementation of theEuropean Water Framework Directive from the BasqueCountry (Northern Spain): A methodological approach.Marine Pollution Bulletin, 48, 209–218.

Borja, A., Galparsoro, I., Solaun, O., Muxika, I., Tello, E.,Uriarte, A., et al. (2006). The European Water FrameworkDirective and the DPSIR, a methodological approach toassess the risk of failing to achieve good ecological status.Estuarine, Coastal and Shelf Science, 66, 84–96.

Borja, A., Valencia, V., Franco, J., Muxika, I., Bald, J.,Belzunce, M. J., et al. (2004b). The Water FrameworkDirective: Water alone, or in association with sediment andbiota, in dtermining quality standards? Marine PollutionBulletin, 49, 8–11.

Bricker, S. B., Ferreira, J. G., & Simas, T. (2003). An integratedmethodology for assessment of estuarine trophic status.Ecological Modelling, 169(1), 39–60.

Chen, Q., Zhang, Y., Ekroos, A., & Hallikainen, M. (2004).The role of remote sensing technology in the E.U. Water

Framework Directive (WFD). Environmental Science &Policy, 7(4), 267–276.

De Jong, F., Bakker, J. F., van Berkel, C. J. M., Dankers, N. M. J.A., Dahl, K., Gätje, C., et al. (1999). Wadden sea qualitystatus report. Wadden sea ecosystem No. 9’, CommonWadden Sea Secretariat, Trilateral Monitoring and assess-ment Group, Quality Status Report Group. Wilhelmshaven,Germany, p. 259.

De Jonge, V. N., Elliott, M., & Brauer, V. S. (2006). Marinemonitoring: Its shortcomings and mismatch with the E.U.Water Framework Directive’s objectives. Marine PollutionBulletin, 53, 5–19.

Ecology and Oceanography of Harmful Algal Blooms (ECO-HAB), undated. http://www.cop.noaa.gov/stressors/extremeevents/hab/current/fact-ecohab.html.

Ecology and Oceanography of Harmful Algal Blooms(ECOHAB), undated. http://www.whoi.edu/ecohab/.

Dworak, T., Gonzalez, C., Laaser, C., & Interwies, E. (2005).The need for new monitoring tools to implement theWFD. Environmental Science & Policy, 8, 301–306.

Elliott, M. (2000). The role of the DPSIR approach andconceptual models in marine environmental management:An example for offshore wind power. Marine PollutionBulletin, 44, iii–vii.

European Community (2000). Directive of the EuropeanParliament and of the Council 2000/60/EC, establishing aframework for community action in the field of waterpolicy, p. 62.

European Commission (2003). Monitoring under the WaterFramework Directive. European Commission, CIS Guid-ance document N° 7, p. 153 http://forum.europa.eu.int/Public/irc/env/wfd/library?l=/framework_directive/guidance_documents&vm=detailed&sb=Title.

European Commission (2006). Common implementationstrategy guidance documents, http://forum.europa.eu.int/Public/irc/env/wfd/library?l=/framework_directive/guidance_documents&vm=detailed&sb=Title.

EUROHAB. http://www.nioz.nl/nioz_nl/e55082b2e503bcf730ee326e48126f2a.php.

Ferreira, J. G., Bettencourt, A., Bricker, S. B., Marques, J. C.,Melo, J. J., Newton, A., et al. (2005). Monitoring plan forwater quality and ecology of Portuguese transitional andcoastal waters. Development of guidelines for the appli-cation of the European Union Water Framework Directive,http://www.monae.org.

Ferreira, J. G., Nobre, A. M., Simas, T. C., Silva, M. C., Newton,A., Bricker, S. B., et al. (2006). A methodology for defininghomogeneous water bodies in estuaries – Application to thetransitional systems of the E.U. Water Framework Direc-tive. Estuarine, Coastal and Shelf Science, In Press.

Ferreira, J. G., Wolff, W. J., Simas, T. C., & Bricker S. B. (2005).Does biodiversity of estuarine phytoplankton depend onhydrology? Ecological Modelling, 187(4), 513–523.

Fraga, S., Anderson, D. M., Bravo, I., Reguera, B., Steidinger, K.A., & Yentsch, C. M. (1988). Influence of upwelling relax-ation on dinoflagellates and shellfish toxicity in Ria de Vigo,Spain. Estuarine, Coastal and Shelf Science, 27(4), 349–361.

Friberg, N., Baattrup-Pedersen, A., Lauge Pedersen, M., &Skriver, J. (2005). The new Danish stream monitoringprogramme (Novana) – preparing monitoring activities for


http://www.bbep.org/ccmp.htm

http://www.ecowin.org/ticor

http://www.bequalm.org/

http://www.cop.noaa.gov/stressors/extremeevents/hab/current/fact-ecohab.html


http://www.whoi.edu/ecohab/

http://forum.europa.eu.int/Public/irc/env/wfd/library?l=/framework_directive/guidance_documents&vm=detailed&sb=Title






http://www.nioz.nl/nioz_nl/e55082b2e503bcf730ee326e48126f2a.php

http://www.nioz.nl/nioz_nl/e55082b2e503bcf730ee326e48126f2a.php

http://www.monae.org

the Water Framework Directive era. Environmental Mon-itoring and Assessment, 111(1–3), 27–42.

Gibson, G. R., Bowman, M. L., Gerritsen, J., & Snyder, B. D.(2000). Estuarine and coastal marine waters: Bioassess-ment and biocriteria technical guidance. EPA822-B-00-024. U.S. Environmental Protection Agency, Office ofWater, Washington, DC.

Goldberg, E. D., Bowen, V. T., Farrington, J. W., Harvey, G.,Martin, J. H., Parker, P. L., et al. (1978). The musselwatch. Environmental Conservation, 5, 101–125.

Joyce, L. B., & Pitcher, G. C. (2004). Encystment ofZygabikodinium lenticulatum (Dinophyceae) during asummer bloom of dinoflagellates in the southern Benguelaupwelling system. Estuarine, Coastal and Shelf Science,59(1), 1–11.

McGillicuddy Jr., D. J., Anderson, D. M., Solow A. R., &Townsend, D. W. (2005). Interannual variability ofAlexandrium fundyense abundance and shellfish toxicityin the Gulf of Maine. Deep Sea Research Part II: TopicalStudies in Oceanography, 52(19–21), 2843–2855.

MEMG, Marine Environment Monitoring Group (2004). UKNational Marine Monitoring Programme – Second Report(1999–2001). Centre for Environment, Fisheries & Aqua-culture Science, CEFAS, p. 136.

Miller, P. I., Shutler, J. D., Moore, G. F., & Groom, S. B.(2005). SeaWiFS discrimination of harmful algal bloomevolution. International Journal of Remote Sensing, 27(11), 2287–2301.

Mostert, E. (2003). The European Water Framework Directiveand water management research. Physics and Chemistry ofthe Earth, 28(12–13), 523–527.

National Oceanic and Atmospheric Administration (NOAA)(2004). Monitoring data – benthic surveillance, NOAANCCOS CCMA. http://www8.nos.noaa.gov/cit/nsandt/download/bs_monitoring.aspx.

National Oceanic and Atmospheric Administration (NOAA)(2005a). Ecology and Oceanography of Harmful AlgalBlooms (ECOHAB), NOAA NCCOS CSCOR, http://www.cop.noaa.gov/stressors/extremeevents/hab/current/fact-ecohab.html.

National Oceanic and Atmospheric Administration (NOAA)(2005b). Monitoring and Event Response for HarmfulAlgal Blooms (MERHAB), NOAA NCCOS CSCOR,http://www.cop.noaa.gov/stressors/extremeevents/hab/cur-rent/fact-merhab.html.

Nobre, A. M., Ferreira, J. G., Newton, A., Simas, T., Icely, J.D., & Neves, R. (2005). Management of coastal eutrophi-cation: Integration of field data, ecosystem-scale simu-lations and screening models. Journal of Marine Systems,56(3/4), 375–390.

OSPAR Commission for the Protection of the Marine Environ-ment of the North–East Atlantic (1999). Guidelines. http://www.ospar.org/asp/ospar/dra.asp?id=3.

OSPAR Commission for the Protection of the Marine Environ-ment of the North–East Atlantic (2004). OSPAR Strategyfor the Joint Assessment and Monitoring Programme(JAMP). http://www.ospar.org/eng/html/strategies/strategy-06.htm.

Salas, F., Neto, J. M., Borja, A., & Marques, J. C. (2004).Evaluation of the applicability of a marine biotic index to

characterize the status of estuarine ecosystems: The caseof Mondego estuary (Portugal). Ecological Indicators, 4,215–225.

Savina, M., & Pouvreau, S. (2004). A comparative ecophysi-ological study of two infaunal filter-feeding bivalves:Paphia rhomboïdes and Glycymeris glycymeris. Aquacul-ture, 239(1–4), 289–306.

Smayda, T. J. (1997). Harmful algal blooms: Their ecophysi-ology and general relevance to phytoplankton blooms inthe sea. Limnology and Oceanography, 42, 1137–1153.

Tillamook Bay National Estuary Program (1999). Restoring thebalance – comprehensive conservation and managementplan for Tillamook Bay, Oregon. Tillamook Bay NationalEstuary Program, Garibaldi, Oregon, completed underCooperative Agreement CE-980127-01 with the U.S.Environmental Protection Agency. http://www.tbnep.org/reports/ccmp.html.

Tillamook Bay National Estuary Project (2005). http://www.co.tillamook.or.us/gov/Estuary/homepage.htm.

Turner, R. K., Adger, W. N., & Lorenzoni, I. (1998). Towardsintegrated modelling and analysis in coastal zones:Principles and practices. LOICZ Reports and studiesN° 11, iv+122 pp. LOICZ IPO, Texel, Netherlands.

United States Environmental Protection Agency (USEPA)(2000). Quality manual for environmental programs.Office of Environmental Information, Quality Staff,5360A1, http://www.epa.gov/quality/qs-docs/5360.pdf.

United States Environmental Protection Agency (USEPA)(2001a). Requirements for quality assurance project plans.Office of Environmental Information, EPA QA/R-5, Rept.EPA/240/B-01/003, Office of Environmental Information,Washington, DC. p. 40. http://www.epa.gov/quality/qs-docs/r5-final.pdf.

United States Environmental Protection Agency (USEPA)(2001b). National Coastal Condition Report (2001) down-load site. http://www.epa.gov/owow/oceans/nccr/downloads.html.

United States Environmental Protection Agency (USEPA)(2002). Guidance for quality assurance project plans formodeling. Office of Environmental Information, EPA QA/G-5M, http://www.epa.gov/quality/qs-docs/g5m-final.pdf.

United States Environmental Protection Agency (USEPA)(2003a). Elements of a state water monitoring and assess-ment program. Assessment and Watershed ProtectionDivision. Office of Wetlands, Oceans and Watershed. U.S. Environmental Protection Agency. EPA 841-B-03-003.p. 24. http://www.epa.gov/owow/monitoring/elements/.

United States Environmental Protection Agency (USEPA)(2003b). Guidance for 2004 Assessment, Listing andReporting Requirements Pursuant to Sections 303(d) and305(b) of the Clean Water Act. United States Environ-mental Protection Agency, Office of Water, Office ofWetlands, Oceans, and Watersheds.

Vincent, C., Heinrich, H., Edwards, A., Nygaard, K., &Haythornthwaite, J. (2003). Guidance on typology, classi-fication and reference conditions for transitional andcoastal waters, European Commission, report of CISWG2.4 (COAST), p. 119 http://forum.europa.eu.int/Public/irc/env/wfd/library?l=/framework_directive/guidance_documents&vm=detailed&sb=Title.


http://www8.nos.noaa.gov/cit/nsandt/download/bs_monitoring.aspx

http://www8.nos.noaa.gov/cit/nsandt/download/bs_monitoring.aspx




http://www.cop.noaa.gov/stressors/extremeevents/hab/current/fact-merhab.html

http://www.cop.noaa.gov/stressors/extremeevents/hab/current/fact-merhab.html

http://www.ospar.org/asp/ospar/dra.asp?id=3

http://www.ospar.org/asp/ospar/dra.asp?id=3

http://www.ospar.org/eng/html/strategies/strategy-06.htm

http://www.ospar.org/eng/html/strategies/strategy-06.htm

http://www.tbnep.org/reports/ccmp.html

http://www.tbnep.org/reports/ccmp.html

http://www.co.tillamook.or.us/gov/Estuary/homepage.htm

http://www.co.tillamook.or.us/gov/Estuary/homepage.htm

http://www.epa.gov/quality/qs-docs/5360.pdf

http://www.epa.gov/quality/qs-docs/r5-final.pdf

http://www.epa.gov/quality/qs-docs/r5-final.pdf

http://www.epa.gov/owow/oceans/nccr/downloads.html

http://www.epa.gov/owow/oceans/nccr/downloads.html

http://www.epa.gov/quality/qs-docs/g5m-final.pdf

http://www.epa.gov/owow/monitoring/elements/




Wang, S., Hong, H., & Wang, X. (2005). Bioenergeticresponses in green lipped mussels (Perna viridis) asindicators of pollution stress in Xiamen coastal waters,China. Marine Pollution Bulletin, 51(8–12), 738–743.

Wätzold, F., & Schwerdtner, K. (2005). Why be wasteful whenpreserving a valuable resource? A review article on thecost-effectiveness of European biodiversity conservationpolicy. Biological Conservation, 123(3), 327–338.

Weiers, S., Bock, M., Wissen, M., & Rossner, G. (2004).Mapping and indicator approaches for the assessment ofhabitats at different scales using remote sensing andGIS methods. Landscape and Urban Planning, 67(1–4),43–65.

WGBEC (2004). Report of the Working Group on Biologicaleffects of contaminants, Ostend, 22–26 March 2004, ICESCM 2004/E:04.


monitoring of coastal and transitional waters under the e ... · water, and coastal and...

Documents