Embed Size (px)
Citation preview
F A C U L T Y O F S C I E N C E U N I V E R S I T Y O F C O P E N H A G E N
Ph.D. thesis Jesper H. Andersen
Ecosystem-Based Management of Coastal Eutrophication Connecting Science, Policy and Society
Submitted: 19/03/2012
1
F A C U L T Y O F S C I E N C E
U N I V E R S I T Y O F C O P E N H A G E N
Ph.D. thesis
Jesper H. Andersen
Ecosystem-Based Management of
Coastal Eutrophication
Connecting Science, Policy and Society
Submitted: 19/03/2012
2
Ph.D. thesis Name of department: Department of Biology Section: Section for Aquatic Biology Author: Jesper H. Andersen Title / Subtitle: Ecosystem-Based Management of Coastal Eutrophication.
Connecting Science, Policy and Society. Topic: This thesis focuses on Ecosystem-Based Management (EBM) of
coastal eutrophication. Special attention is put on connections be-tween science and decision-making in regard to development, im-plementation and revision of evidence-based nutrient management strategies. Two strategies are presented and analysed: the Danish Action Plans on the Aquatic Environment and the eutrophication segment of the Baltic Sea Action Plan. Similarities and differences are discussed and elements required for making nutrient manage-ment strategies successful are suggested.
Key words: Eutrophication, marine, Danish waters, Baltic Sea, action plans, nu-
trient management strategies, adaptive management, ecosystem-based management, monitoring, assessment.
Number of pages: 54 + annexes Submitted: 19 March 2012
3
List of Contents
Preface 4
Danish Summary (Resumé) 6
Summary 8
Abbreviations 12
1: The Danish Action Plans on the Aquatic Environment and the Baltic Sea
Action Plan: Two Successful Nutrient Management Strategies?
13
2: Similarities and Differences Between the Danish Action Plans on the
Aquatic Environment and the Baltic Sea Action Plan
31
3: Beyond Action Plans and Directives: Perspectives for the Future
42
4: Conclusions: What Makes a Nutrient Management Strategy Successful? 45
5: References 48
Annex 1: Abstract 55
Annex 2: Nutrient Discharges and Losses in Denmark 1989-2008 56
Annex 3: Curriculum Vitae for Jesper H. Andersen 60
Annex 4: Manuscripts 74
4
Preface
This thesis addresses cultural eutrophication and the management of its causes.
The objectives of the thesis are:
1. To present and evaluate data from two apparently successful evidence-based nutrient man-
agement strategies: the Danish Action Plans on the Aquatic Environment and the eutrophica-
tion segment of the HELCOM Baltic Sea Action Plan.
2. To analyse and discuss suitable management strategies based on the two case studies.
Eutrophication itself cannot be managed; only the human activities leading to nutrient enrich-
ment and eutrophication are within reach of control. Sometimes we incorrectly speak about man-
agement of eutrophication, when we mean development and implementation of strategies to
change human behaviour with an ultimate aim of reducing direct discharges, diffuse losses
and/or emissions (to the atmosphere) of nutrients to the aquatic environment. Special focus is put
on those links between science and decision-making processes that make nutrient management
strategies effective.
The thesis is structured in the following way:
An interdisciplinary and cross-cutting synthesis with focus on the long-term implementation
and development of two nutrient management strategies, and
An annex including the peer-reviewed papers on which this cross-cutting synthesis is based.
The thesis is founded on the following publications:
1. Conley, D.J., S. Markager, J.H. Andersen, T. Ellermann & L.M. Svendsen, 2002: Coastal
Eutrophication and the Danish National Aquatic Monitoring and Assessment Program. Estu-
aries 25:848-861.
2. Andersen, J.H., D.J. Conley & S. Hedal, 2004: Palaeo-ecology, reference conditions and
classification of ecological status: the EU Water Framework Directive in practice. Marine
Pollution Bulletin 49:283-290.
3. Andersen, J.H., L. Schlüter & G. Ærtebjerg, 2006: Coastal eutrophication: Recent develop-
ments in definitions and implication for monitoring strategies. Journal of Plankton Research
28(7):621-628.
4. Carstensen, J., D.J. Conley, J.H. Andersen & G. Ærtebjerg, 2006: Coastal eutrophication
and trend reversal: A Danish case study. Limnology & Oceanography 51(1-2):398-408.
5. Andersen, J.H., & D.J. Conley, 2009: Eutrophication in coastal marine ecosystems: towards
better understanding and management strategies. Hydrobiologia 621(1):1-4.
6. Andersen, J.H., C. Murray, H. Kaartokallio, P. Axe & J. Molvær, 2010: A simple method
for confidence rating of eutrophication status classifications. Marine Pollution Bulletin
60:919-924.
5
7. Andersen, J.H., 2010: Eutrophication. Baltic Sea Environmental Proceedings 122:16-17. In:
HELCOM, 2010: Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment
2003-2007. Baltic Sea Environment Proceedings 122. 63 pp.
8. Andersen, J.H., P. Axe, H. Backer, J. Carstensen, U. Claussen, V. Fleming-Lehtinen, M.
Järvinen, H. Kaartokallio, S. Knuuttila, S. Korpinen, A. Kubiliute, M. Laamanen, E. Lysiak-
Pastuszak, G. Martin, F. Møhlenberg, C. Murray, G. Nausch, A. Norkko & A. Villnäs, 2011:
Getting the measure of eutrophication in the Baltic Sea: towards improved assessment prin-
ciples and methods. Biogeochemistry 106:137-156.
9. Korpinen, S., L. Meski, J.H. Andersen & M. Laamanen, 2012: Human pressures and their
potential impact on the Baltic Sea ecosystem. Ecological Indicators 15:105-114.
10. Laamanen, M., S. Korpinen, U.-L. Zweifel & J.H. Andersen, in review: Ecosystem health.
Textbook chapter in “Biological Oceanography of the Baltic Sea” (Eds: P. Snoeijs, H. Schu-
bert & T. Radziejewska).
The thesis also draws on information from the peer-reviewed reviewed book:
11. Ærtebjerg, G., J.H. Andersen & O.S. Hansen, 2003: Nutrient and Eutrophication in Danish
Marine Waters. A Challenge to Science and Management. National Environmental Research
Institute. Roskilde. 126 p.
The production of this thesis has been financially supported by DHI (RK 2006-2009). This thesis
would not have been possible without the support of the DHI-NTU Water & Environmental Re-
search Centre in Singapore.
Special thanks are due to Daniel J. Conley for discussion of an earlier version of the thesis and
linguistic corrections as well as to Gunni Ærtebjerg, Jørn Kirkegaard and Mette Olesen.
Thanks are extended to J. Borum, J. Brøgger Jensen, J. Carstensen, U. Claussen, B.W. Hansen,
J.W. Hansen, H. Karup, S. Korpinen, M. Laamanen, J.E. Larsen, J.-M. Leppänen, O. Mark, C.
Murray, F. Møhlenberg, S. Pedersen, J.D. Petersen, J.B. Reker, B. Riemann and A. Stock.
Jesper H. Andersen
19 March 2012
6
Danish Summary (Resumé)
Denne afhandling handler om ’eutrofiering’ og forvaltning af årsagerne til eutrofiering.
Udledning og tab af næringsstoffer til vandmiljøet fører til eutrofiering, et ord som har sine rød-
der i to græske ord: ‘eu’ som betyder ‘godt’ og ‘trope’ betyder ’næret’. Den moderne brug af
ordet ‘eutrofiering’ er relateret til tilførsler og effekter af næringsstoffer i økosystemer, specielt
næringsstofberigelse af akvatiske økosystemer. Det er med de danske kystvande som med men-
nesker, der spiser for godt: for megen mad fører til en dårlig sundhedstilstand. Velkendte symp-
tomer på, at ’patienten’ ikke har det godt er bl.a. algeopblomstring og iltsvind.
Formålet med afhandlingen er udover at præsentere og diskutere principperne for ’adaptiv ma-
nagement’ og ’ecosystem-based management’, at (1) analysere og diskutere to videnbaserede
forvaltningsstrategier, at (2) identificere hvad der som minimum skal til, for at strategiske miljø-
handleplaner kan gennemføres med succes.
Strategiske miljøhandleplaner og egentlige forvaltningsplaner med fokus på næringsstoffer og
næringsstofforurening i vandmiljøet, skal som udgangspunkt inkludere samtlige menneskelige
aktiviteter og kilder, hvorfra der bliver udledt eller ’tabt’ næringsstoffer. Disse indbefatter land-
brug, industri, husholdninger og energiproduktion og -forbrug.
To forvaltningsplaner, som vurderes at være bedste praksis i forhold til udvikling og gennemfø-
relse af strategi- og forvaltningsplaner til nedbringelse af tilførslerne af næringsstoffer til vand-
miljøet og til begrænsning af eutrofieringseffekterne i hhv. danske farvande og Østersøen, analy-
seres. Den danske forvaltningsplan er vandmiljøplanerne fra 1987, 1998 og 2004, som altoverve-
jende er et eksempel på ’adaptiv management’. Den anden forvaltningsplan er Østersøhandlings-
planen fra 2007, der er baseret på ‘ecosystem approach to management of human activities’, der i
princippet er identisk med ’ecosystem-based management’.
Afhandlingen beskriver de faktorer, der er kritiske, hvis en strategi eller forvaltningsplan på sigt
skal blive en succes og føre til væsentlige reduktioner af udledninger og tab af næringsstoffer til
vandmiljøet. Nøglefaktorer er progressive reduktioner af udledningerne af næringsstoffer, gene-
relt både kvælstof (N) og fosfor (P), landsdækkende overvågning af tilførsler til og effekter af
næringsstoffer i vandmiljøet samt politisk vilje til at fastholde oprindeligt fastsatte mål.
Politisk fokus og vilje til handling på området er afgørende. Om reduktionsmålene er tilstrække-
lige, er tilsyneladende ikke afgørende for, om en strategi eller forvaltningsplan bliver en succes –
i hvert fald ikke så længe både analyse og rapportering af den gennemførte overvågning og re-
gelmæssig evaluering af udviklingen i tilførslerne af næringsstoffer har en central plads i strate-
gien. Overvågningen skal føre til årlig opgørelse af tilførslerne og årlige vurderinger af miljøtil-
standen. Vurderingerne skal i sagens natur udarbejdes af faglige institutioner og evalueres af
beslutningstagere eller den statslige administration, der traditionel understøtter politikerne i fag-
ligt komplicerede spørgsmål.
Der er flere forskellige måder, hvorpå økosystem-baserede forvaltningsstrategier og -planer kan
udarbejdes. Den enkelte plan skal tage udgangspunkt i det pågældende områdes karakteristika.
Forvaltningsstrategier og -planer vil desuden variere afhængigt af den anvendte lovgivning og
hvilke myndigheder, der er involveret. De to forvaltningsplaner har mange lighedspunkter, men
er på en række centrale punkter væsentligt forskellige, bl.a. med hensyn til geografisk dækning
7
og politisk ophæng: Den ene er national og gennemføres med lovgivning. Den anden er interna-
tional og gennemført med en politisk aftale.
Afhandlingen konkluderer at følgende elementer er afgørende for at forvaltningsstrategier og
-planer kan blive gennemført med succes:
Miljø- og reduktionsmålene skal være klare for alle parter.
Effekter og udledninger skal løbende overvåges.
Strategien eller planen skal indeholde: 1) målfastsættelse, 2) virkemidler, 3) overvågning og
4) evalueringsfase.
Den politiske vilje skal være til stede.
Et forslag til hvordan disse elementer kan kobles, er illustreret nedenfor:
Konceptuel model for tilrettelæggelse og gennemførelse af en økosystem-baseret forvaltningsstrategi. Bemærk at relevante eutrofieringseffekter er indeholdt, både forhøjede koncentrationer af næringsstoffer, direkte effekter (eksempelvis primærproduktion og klorofyl-a-koncentrationer) og indirekte effekter (for eksempel iltsvind og ændring i mængde og udbredelse af bundlevende dyr og planter).
Fremtidige forvaltningsstrategier og -planer vil ikke skulle opbygges fra grunden af, men bygge
på det eksisterende. Første trin vil være at tage det bedste fra ’adaptive management’ og kombi-
nere det med det bedste fra ’ecosystem-based management’. Konkret vil det væsentligst omfatte
processen fra førstnævnte og den økosystem-baserede tilgang fra sidstnævnte. Den måske mest
kritiske faktor med hensyn til, om fremtidige forvaltningsplaner kan føre til en god miljøtilstand,
er, at de ikke bliver udarbejdet og gennemført for snævert, både hvad angår fokus og person-
kreds. Fremtidige forvaltningsplaner bør derfor være videnbaserede, involvere alle relevante
interessenter og nyde opbakning af en vedvarende politisk vilje til at begrænse tilførslerne af
næringsstoffer til vandmiljøet.
8
Summary
This thesis concerns eutrophication and the management of human activities resulting in nutrient
enrichment and the biological effects on aquatic ecosystems.
The term ‘eutrophication’ (noun) has its root in two Greek words: ‘eu’ which means ‘well’ and
‘trope’ which means ’nourishment’. The modern use of the word eutrophication is related to high
inputs and effects of inorganic nutrients in ecosystems, especially over-enrichment of aquatic
ecosystems.
Management is basically the process of getting people together to accomplish desired goals and
objectives. The verb ‘manage’ comes from the Italian ‘maneggiare’ (to handle), which originally
derives from the Latin ‘manus’ (hand). In the context of eutrophication, management is about
setting up a strategy for control of human activities resulting in discharges (direct sources), loss-
es (diffuse sources, e.g. from agriculture) and emissions (to the atmosphere) of nitrogen, phos-
phorus and organic matter to the aquatic environment.
An adaptive nutrient management strategy (NMS) should include the following elements: Prob-
lem identification and four phases focusing on planning, acting, checking and evaluation. The
papers on which this thesis is based upon addresses all of these five elements. (See Table 1).
Paper no. Context Planning Acting Checking Evaluation
1. Conley et al. (2002) (x) (x) X
2. Andersen et al. (2004) (x) X
3. Andersen et al. (2006) X (x)
4. Carstensen et al. (2006) (x) (x) X (x)
5. Andersen & Conley (2009) (x) X
6. Andersen et al. (2010) (x) X (x)
7. Andersen (2010) (x) X (x)
8. Andersen et al. (2011) (x) X
9. Korpinen et al. (2012) (x) X
10. Laamanen et al. (submitted) (x) X (x)
‘X’ = the paper has a direct focus upon this NMS phase; ‘(x)’ = the paper has an indirect focus.
Understanding the context of eutrophication is important both from a scientific point of view,
since both definitions and conceptual understanding are constantly developing, and from an
implementation of nutrient management strategies. If decision-makers are not informed or do not
understand the concept of eutrophication then management is a difficult task. Despite a wide-
spread common European understanding of causes and effects of eutrophication, there is no
mutually agreed definition of coastal eutrophication. However, within the European Union (EU)
there has been a sound tradition of focusing the measures on the sources causing eutrophication
(Elliot et al. 1999, Elliot & de Jonge 2002). Consequently, eutrophication has been defined in
relation to sources and/or sectors. For example, in the EU Urban Waste Water Treatment
Directive, eutrophication has been defined as “the enrichment of water by nutrients, especially
nitrogen and/or phosphorus, causing an accelerated growth of algae and higher forms of plant
life to produce an undesirable disturbance to the balance of organisms present in the water and to
the quality of water concerned” (Anon. 1991a). The EU Nitrates Directive has an almost
identical definition specifically emphasising losses of nitrates from agriculture (Anon. 1991b).
9
Nixon (1995) defined eutrophication as “an increase in the rate of supply of organic matter to an
ecosystem”. This definition is short and emphasizes that eutrophication is a process, not a trophic
state. Nixon also noted that various factors may increase the supply of organic matter to coastal
systems, but the most common is clearly nutrient enrichment. The supply of organic matter to an
ecosystem is not restricted to pelagic primary production, even though such an interpretation
makes the definition operational. The supply of organic matter to a system includes primary
production of higher plants and benthic microalgae as well as inputs of organic matter from adja-
cent waters or from land via rivers or point sources. Having such a broad interpretation of the
term ‘supply’ makes the definition difficult to use in a monitoring and management context.
Eutrophication and definition(s) of eutrophication are widely discussed (Jørgensen & Richardson
1996). The most common use of the term is related to inputs of mineral nutrients, in particular
nitrogen and phosphorus, to specific waters. Consequently, eutrophication deals with both the
process and the associated effects of nutrient enrichment and natural versus cultural
eutrophication. Despite the definitions in existing European directives, the implementation of the
EU Water Framework Directive (WFD) revealed a need for a common understanding and
definition of eutrophication as well as stronger co-ordination between directives dealing directly
or indirectly with eutrophication. Hence, the European Commission convened a process aiming
for a development of a pan-European conceptual framework for eutrophication assessment in the
context of all European waters and policies (Anon. 2009a). This process did not lead to a
common European definition of eutrophication, but it revealed that if ‘undesirable disturbance’ is
understood as ‘unacceptable deviation from reference conditions’, the pan-European definition
will be coherent with the normative definitions sensu the WFD (Andersen et al. 2006).
Accepting this, a pan-European definition of eutrophication, would be:
“the enrichment of water by nutrients, especially nitrogen and/or phosphorus, and orga-
nic matter, causing an increased growth of algae and/or higher forms of plant life to pro-
duce an unacceptable deviation in structure, function and stability of organisms present
in the water and to the quality of water concerned, compared to reference conditions”.
The suggested definition includes causative factors (nutrient enrichment), primary effects
(increased growth) and secondary effects (sometime referred to as ‘undesirable disturbance’).
However, it also is a matter of interpretation, in particular in regard to what an ‘acceptable devia-
tion’ is.
In addition, the definition enables classification of ‘eutrophication status’. Using the definition as
a basic assessment principle, an eutrophication quality objective or target (EutroQO) is defined
as an indicator with an acceptable deviation (AcDev) from the reference condition (RefCon),
EutroQO = RefCon ± AcDev (Andersen et al. 2004, Andersen et al. 2011). As an additional
feature, the definition also acknowledges that eutrophication has both quantitative and qualitative
perspectives, an aspect not included in Nixon’s definition.
The setting of science-based eutrophication quality objectives (or targets) is a prerequisite for
ecosystem-based management. These target setting principles used in Europe are commonly
based of information on RefCon and setting of an AcDec from RefCon. The concept originates
from the Water Framework Directives and is described and demonstrated by Andersen et al.
(2004). The strength of the concept is that it is operational and that the data used derived by
science-based process. The weakness is that is allows for expert judgement, e.g. in regard to the
10
setting of AcDev, and thus potentially a weakening of the scientific basis. A specific problem is
related to natural variability and it potential influence of the assessment of eutrophication status
(Andersen et al. 2004).
Before implementing a nutrient management strategy, having a complete overview of the human
activities and pressures is necessary to focus on the activities resulting in impaired conditions.
An example can be found in Korpinen et al. (2012), where cumulative pressures and impacts in
the Baltic Sea region have been estimated. The estimate is based on the mapping of human
activities, maps of key ecosystem components and expert judgement of the impact of a specific
pressure upon a specific ecosystem component. A matrix is established and from it, the dominant
pressures in the Baltic Sea were estimated to be: (1) nutrient enrichment, (2) fishing activity, (3)
input of contaminants, and (4) physical modification. This study is the first ever assessment of
cumulative pressures and impacts for a regional sea, and is a useful tool for documenting the
causes of impaired conditions as well as targeting of measures, regionally and sub-regionally.
The targets of nutrient management can in principle be established in two ways: (1) the tradition-
al way where load reduction targets are agreed upon, and (2) a more modern way where Eu-
trophication Quality Objectives (EutroQO’s) are established and the critical loads matching the
EutroQO’s are calculated. Two different case studies are analysed in this thesis. The Danish Ac-
tion Plans on the Aquatic Environment have a strong focus on load reduction targets for agri-
cultural discharges and losses as well as discharges from urban water treatment plants and indus-
tries with separate discharge (Conley et al. 2003, Carstensen et al. 2009, Andersen & Conley
2009). The HELCOM Baltic Sea Action Plan, which is based on an ecological target and subse-
quent calculation of a critical load, represent a more evidence-based way to estimate the load
reductions (Andersen et al. 2011).
A key step in any nutrient management strategy is monitoring for expected improvements in
ecological quality, specifically eutrophication status, in the marine environment. The Danish
Action Plans on the Aquatic Environment included a well-designed monitoring programme for
Danish marine waters (Conley et al. 2002, Ærtebjerg et al. 2003). The data and information orig-
inating from monitoring activities have not only resulted in annually national reports, which have
been used for regular evaluations of the effectiveness of the Danish Action Plan, but also in pa-
pers of eutrophication trends (Carstensen et al. 2006). Both the reductions in loads and the ef-
fects of the loads reductions in Danish coastal waters are well documented: (1) inputs have de-
creased significantly, both for nitrogen and phosphorus, (2) nutrient concentrations have de-
creased significantly, (3) primary productivity and phytoplankton biomass have decreased as
well, and (4) benthic communities have in some areas improved their ecological status.
The work on assessing eutrophication in Danish marine waters and the Baltic Sea has lead to
important advances in our understanding. For decades, eutrophication assessments have focused
on state for a given indicator supplemented by temporal trend assessment for individual
indicators. Recently, multi-metric indicator-based assessment tools are emerging (Andersen
2010, Andersen et al. 2011). With the development of the HELCOM Eutrophication Assess-
ment Tool (HEAT) (Andersen et al. 2011), status assessments can now be supplemented with a
simple estimate of confidence (Andersen et al. 2010). The approach to solve a statistical
challenge in a non-statistical way is based on expert judgement of the confidence of information
in regard to RefCon, AcDec and observations of the state. This information is combined for each
indicator and integrated into an overall estimate of confidence. Information in regard to confi-
11
dence estimates is useful for setting up evidence-based nutrient management strategies, but also
essential when redesigning monitoring programmes.
Based on the lessons from the Danish Action Plans on the Aquatic Environment and the
HELCOM Baltic Sea Action Plan, the following DO’s and DON’T’s of evidence-based nutrient
management strategies can be made:
DO understand that ecosystem-based management is adaptive and science-based.
DON’T assume that decisions can not be taken because of incomplete knowledge and
uncertainty.
DO evidence-based target setting and exhaustive planning, the latter involving decision-
makers, authorities and all stakeholders.
DON’T wait for perfection and all-inclusive ecosystem understanding.
DO a full execution of the plan.
DON’T rely on voluntary agreements or guidelines.
DO monitoring with ecologically relevant resolution in time and space.
DON’T underestimate resources needed for sampling, quality assurance, analysing data and
reporting.
DO regular evaluations in regard to the progress of the nutrient management strategy.
DON’T disregard the advantages of a dual monitoring strategy focusing on both nutrient
inputs as well as ecological responses to lowered nutrient inputs.
An important lesson learned from the Danish Action Plans on the Aquatic Environment and the
HELCOM Baltic Sea Action Plan is that decisions are often made in short windows of
opportunity. It is critical to prepare for those brief moments where decisions and actions can be
taken. Preparation of decision support systems and determining the best possible scientific basis
for decision-making can provide the scientific basis for actions to be implemented.
Perhaps the most important lesson is that time is needed before the effects of changes in human
behaviour can be observed in nutrient inputs and, eventually, in the ecological quality of the
marine environment. It would, therefore, be prudent to ask if we, within a decade or two, can
expect to have a marine environment not affected by eutrophication as required by national and
international processes, e.g. the Danish Action Plan on the Aquatic Environment, the EU Water
Framework Directive, the EU Marine Strategy Directive and the HELCOM Baltic Sea Action
Plan. There are a number of factors underlying the slow response of ecosystems, e.g. delays in
nutrient inputs from fields to streams and rivers caused by retention in the soil. There is a grow-
ing recognition that the recovery trajectories differ from the well known degradation trajec-
tories (Duarte et al. 2009; Laamanen et al. submitted). Another challenge is a shifting baseline
caused by increasing temperatures, resulting in a situation where loads of nutrients probably
have to be reduced more than estimated in a situation with stable temperatures (Laamanen et al.
submitted). Apparently, we face two counteracting process, one where nutrient loads are pro-
gressively reduced, and one where sea temperatures are rising. The prospects in regard to the
long-lasting eutrophication crisis are not good due to the lack of political will to act, a fact being
sustained by the current financial crisis.
12
Abbreviations
AcDev Acceptable deviation
AM Adaptive Management
APAE Action Plan on the Aquatic Environment
BSAP HELCOM Baltic Sea Action Plan
COMBINE Cooperative Monitoring in the Baltic Marine Environment
DAMP Danish Aquatic Monitoring Programme
EBM Ecosystem-Based Management
EC European Community
EEZ Exclusive Economic Zone
EU European Union
EutroQO Eutrophication Quality Objective
HEAT The HELCOM Eutrophication Assessment Tool
HELCOM Helsinki Commission
NMS Nutrient Management Strategy
MSFD Marine Strategy Framework Directive
N Nitrogen
ND Nitrates Directive
NGO Non-Governmental Organisation
NOVA Nationalt program for overvågning af vandmiljøet (DAMP 1998-2003)
NOVANA Nationalt overvågningsprogram for vandmiljøet og naturen (DAMP 2004-2015)
NPo Nitrogen, phosphorus and organic matter
OSPAR Oslo and Paris Commissions
P Phosphorus
PACE Plan, act, check and evaluate
RBMP River Basin Management Plan
RefCon Reference conditions
RT Reduction target
TN Total nitrogen
TL Target load
TP Total phosphorus
UWWTD Urban Waste Water Treatment Directive
UWWTP Urban waste water treatment plant
WFD Water Framework Directive
13
1: The Danish Action Plans on the Aquatic
Environment and the Baltic Sea Action Plan:
Two Successful Nutrient Management Strategies?
Cultural eutrophication of coastal waters has been recognised as a growing global problem for
more than three decades. Many resources have been allocated for eutrophication research and
eutrophication mitigation around the world, especially in Europe and North America. The caus-
es, processes, and effects of eutrophication are well documented (e.g. Cloern 2001, Kononen &
Bonsdorff 2001, Rabalais & Nixon 2002, Bachmann et al. 2006, Diaz & Rosenberg 2008, An-
dersen & Conley 2009). However, very few examples of successful nation-wide or regional
nutrient management strategies (NMS) are published (e.g. Carstensen et al. 2006, Kronvang et
al. 2008). This raises a series of questions: (1) Do we have a common conceptual understanding
of what eutrophication and nutrient management strategies are about? (2) For decades, NMS’s
have been planned and implemented, but why have plans in general not resulted in significant
improvements? (3) Do we have to wait for the effects of already implemented actions or is it
possible that we have a structural defect preventing our plans from succeeding?
A hypothesis in this thesis is that any successful NMS or action plan is characterised by the con-
fluence of the following four steps: (1) a politically agreed plan including objectives and targets,
(2) implementation of measures, (3) monitoring activities including publication of assessments,
and (4) appropriate feedback loops from monitoring and assessment to the political level (back to
step 1). Direct testing of this working hypothesis is not possible. Instead this thesis analyses and
discusses two apparently successful nutrient management strategies, (1) a national action plan
based on Adaptive Management (AM), and (2) a trans-national action plan, aiming to be based
on the principles of ecosystem-based management (EBM).
The two action plans differ substantially. The one based on the AM approach has a long history,
while the one based on the EBM approach has had a long prologue, but is in the early phases of
its implementation. Both plans are believed to be representative in regard to AM and EBM, re-
spectively, and they are analysed and discussed with the aim of highlighting which factors an
evidence-based NMS might include in order to be successful. The first example of an apparently
successful nutrient management strategy are the Danish Action Plans on the Aquatic Environ-
14
ment (APAE), where the first of, to date, three consecutive plans was adopted in 1987. The in-
formation about APAE originates from a combination of governmental publications and White
Papers as well as peer reviewed papers, e.g. Iversen et al. (1999), Grant et al. (2006), Carstensen
et al. (2006) and Kronvang et al. (2008). It should be pointed out that APAE 1 does not have any
reference since the first APAE 1 is a combination of a proposal from the Danish Government
(Miljøministeriet 1987) and changes to it decided by a majority of the Danish Parliament (Folke-
tinget 1986-1987). The second nutrient management strategy is the HELCOM Baltic Sea Action
Plan (BSAP) which covers an entire regional sea. The information about the HELCOM BSAP
originates from the BSAP itself (HELCOM 2007) and a suite of peer reviewed papers, e.g. Sav-
chuk & Wulff (2007), Wulff et al. (2007), Backer (2008), Backer & Leppänen (2008) and Back-
er at al. (2009). In addition, information has been extracted from HELCOM (2009). The Danish
marine waters as well as the neighbouring regional seas, the wider North Sea and the Baltic Sea
are shown in Figure 1. It should be noted that the Kattegat and the Danish Straits, being the tran-
sition zone between the North Sea/Skagerrak and the Baltic Sea, as well as the marine waters
around the island of Bornholm, are included in both plans.
Figure 1: Map of the Baltic Sea and North Sea including the transition zone consisting of the Skagerrak
between Denmark, Norway and Sweden, the Kattegat and the Danish Straits (between the main Danish islands west of southern Sweden). EEZ = Exclusive Economic Zone.
15
Other nutrient management plans were considered in the analysis. Based on the available scien-
tific literature and combination of criteria (plans should focus on both point and diffuse sources,
not operate on a local scale, and have been enacted), only the above introduced nutrient man-
agement strategies were selected cf. Table 1.
An additional reason for focusing on the Danish APAE’s and the HELCOM BSAP was that it is
well documented that nutrient enrichment is the key pressure followed by fisheries, inputs of
heavy metal and inputs of persistent organic pollutants (HELCOM 2010, Korpinen et al. 2012).
Similar rankings of cumulative pressures are not available for the other areas where nutrient en-
richment may be an issue1 though it can not be excluded that other pressures are more important.
Table 1: An overview of potential successful nutrient management strategies including criteria for final selection. P = point sources; D = diffuse sources; NAT = national plan or strategy; REG = regional plan or strategy; LOC = local plan or strategy; UW = union-wide; and UWWT = urban waste water treatment. CBA = Chesapeake Bay Agreement; GHAP = Gulf Hypoxia Action Plan; and VLSL = Venice Lagoon Special Law.
Plan/Strategy Adopted Predecessor P / D Scale Enacted Reference
Danish APAE 1987 NPo Action Plan P+D NAT + ATV 1990 HELCOM BSAP 2007 HELCOM 50% P+D REG Indirectly
1 HELCOM 2007
HELCOM 50% 1988 None P+D REG ÷ Laäne et al. 2002 OSPAR 50% 1988 None P+D REG ÷ OSPAR 2008
Chesapeake Bay 2000 CBA 1983, 1987 P+D REG ÷ Bosch et al. 2001 Gulf of Mexico 2008 2011 GHAP P+D REG ÷ Anon. 2008
Venice Lagoon 1992 VLSL 1973 P+D LOC + Suman et al. 2005
EC Nitrates Dir. 1991 None D UW + Anon. 2010 EC UWWT Dir. 1991 None P UW + Anon. 1991b
1: The HELCOM Baltic Sea Action Plan is indirectly enacted via the EC Urban Waste Water Treatment Directive and the EC Nitrates Directive as eight out of nine coastal states are EU Members States. 2: ‘Eutrophication non-problem area’ cf. OSPAR and not designated as a sensitive sensu the UWWTD. 3: The upstream catchment is designated as a ‘nitrogen vulnerable zone’ sensu the Nitrates Directive.
The Danish APAE’s and in particular the HELCOM BSAP are likely to represent best practices
in regard to European nutrient management strategies (Foden et al. 2008). No other successful
national or regional plans have been identified, although the upcoming WFD River Basin Man-
agements Plans (RBMPs) include many features of evidence-based and adaptive nutrient man-
agement strategies (Foden et al. 2008).
1 The Chesapeake Bay, United States may potentially be considered a successful nutrient management strategy
(based on information in Bosch et al. (2001), Kemp et al. (2005), and Bosch (2006)), but detailed information on
cumulative pressures and impacts sensu Halpern et al. (2008) is not available.
16
Having an understanding of the meaning of the terms ‘Adaptive Management’ (AM), and ‘Eco-
system-Based Management’ (EBM) is important for two reasons: first, and from a societal point
of view, because policy drivers and management frameworks are being continuously developed
and updated, and second, because scientists should have knowledge regarding legislative and
political processes to understand that decision-making has to balance recommendations from
scientists with societal needs.
‘Adaptive Management’ (noun) is a
structured, iterative process of best pos-
sible decision making in the face of un-
certainty, seeking to reduce uncertainty
over time via system monitoring. AM is
often characterized as "learning by do-
ing” and depends upon an open manage-
ment process which seeks to include past,
present and future stakeholders, for ex-
ample those sectors discharging, losing
or emitting nutrients to the environment.
Hence, AM is characterised as being both
a social and a scientific process. In its
basic form, AM includes four phases: (1)
a planning phase, (2) an action phase, (3)
a checking phase, and (4) an evaluation
phase. This sequence is on occasion
named PACE (Figure 2).
AM is linked to the ‘Ecosystem Ap-
proach to management of human activi-
ties’ (EA). EA (noun) is defined as ”the
comprehensive integrated management
of human activities based on the best
available scientific knowledge about the
ecosystem and its dynamics, in order to
Figure 2: Conceptual model of the classical Adaptive
Management cycle including five phases: Identification (Do we have a problem?) and the Plan, Action, Check and Evaluate loop, the four last phases sometimes grouped under a PACE heading. Feedback loops exist from evaluation to implementation and planning, the latter including adoption of additional measures. .
17
identify and take action on influences which are critical to the health of marine ecosystems,
thereby achieving sustainable use of ecosystem goods and services and maintenance of ecosys-
tem integrity” (HELCOM & OSPAR 2003). Hence, the Ecosystem Approach can be seen as a
fore-runner of ‘ecosystem-based management’ as indicated by Backer et al. (2009).
‘Ecosystem-Based Management’ (noun) is an integrated approach to management that considers
the entire ecosystem, including humans with the goal of maintaining an ecosystem in a healthy,
productive and resilient condition so that it can provide the services humans want and need
(McLeod et al. 2005). An important element in regard to Ecosystem-Based Management (EBM)
is the term ‘ecosystem’ (noun), which is “a dynamic complex of plant, animal and micro-
organism communities and their non-living environment interacting as a functional unit”, cf. the
UN Convention on Biological Diversity.
EBM differs from current approaches that focus on a single species, sector, activity or concern; it
considers the cumulative impacts of different sectors. Specifically, EBM: (1) emphasizes the
protection of ecosystem structure, functioning, and key processes; (2) focuses on a specific eco-
system and the range of activities affecting it; (3) explicitly accounts for the interconnectedness
within systems, recognizing the importance of interactions between many target species or key
services and other non-target species; (4) acknowledges interconnectivity among systems, such
as between air, land and sea; and (5) integrates ecological, social, economic, and institutional
perspectives, recognising their strong interdependences (Christensen et al. 1996, McLeod et al.
2005). At its core, EBM is about acknowledging linkages between ecosystems and human socie-
ties, economies and institutional systems (McLeod & Leslie 2009).
EBM and AM share a lot of common ground, especially in regard to the Action and Checking
phases. The most prominent differences between EBM and AM are related to the Planning
phase, which in regard to EBM is principally evidence-based since it is system-oriented and
based on the best available knowledge, and to the Evaluation and feedback phase which is a key
phase in the AM process, while it is not being given specific emphasis or is an integrated part of
the current EBM concept. Another key differentiation is that AM generally focuses on sectors
and their pressures, whilst EBM aims on setting ecologically relevant targets.
18
1.1: The Danish Action Plans on the Aquatic Environment
In early autumn of 1986, large parts of the Danish Straits and estuaries were depleted of oxygen.
Danish fishermen showed dead Norwegian lobsters on national television and thereby demon-
strated to the public and politicians that the environmental status of the marine waters of Den-
mark was severely impaired (ATV 1990, Andersen & Carstensen 2011).
The public communication of the dead lobsters is generally considered to be the catalyst of the
Danish Action Plans on the Aquatic Environment (APAE). However, in reality the issue of eu-
trophication emerged gradually during the 1970s, with the Belt Project as one of the activities
launched to document the extent and severity of the effects of nutrient enrichment on marine
waters in Denmark (Ærtebjerg Nielsen et al. 1981).
The Belt Project, taking place 1975-1978, was the first Danish National Marine Research Pro-
gramme focusing on nutrient enrichment and its associated effects. It showed that nutrient con-
centrations, primary production and phytoplankton biomass were increasing and that oxygen
concentrations were decreasing, but also that there were no general problems related to nutrient
enrichment in the open waters (Ærtebjerg Nielsen et al. 1981). This conclusion was questioned,
since a number of incidents occurred in the summer and early autumn of 1981, where fish and
benthic invertebrates were killed by oxygen depletion (Miljøstyrelsen 1984a). When the Belt
Project ended in 1978, parts of it continued as an ongoing activity named the National Marine
Pollution Monitoring Programme. Focus was on monitoring of nutrients, phytoplankton and ox-
ygen in the open parts of the Inner Danish Waters (the Kattegat, the Danish Straits and the south
western part of the Baltic Sea).
The report “Oxygen depletion and fish kills in 1981” (Miljøstyrelsen 1984a), primarily being
based on local monitoring activities, and the NPo White Paper (Miljøstyrelsen 1984b) put focus
on an emerging and extensive problem related to nutrient enrichment and derived consequences
in Danish Marine waters. As a consequence, the Ministry of Environment developed the NPo
Action Plan from 1985 focusing on discharges of nitrogen (N), phosphrous (P) and organic mat-
ter (o) from point sources, including the setting of discharge limits in effluents from urban waste
water treatment plants (Folketinget 1984-1985). Discharges and losses of nutrients from agricul-
ture were not taken into consideration, because of uncertainties in regard to different sources and
pathways. Linked to the plan was a national research programme, the NPo Research Programme,
19
focusing on discharges to groundwater and fresh and marine surface waters, the effects of these
discharges, and possible alleviating activities.
In 1986, a few days after the lobsters were killed by hypoxia, the Danish Society for Nature Con-
servation (DN), the largest nature conservation and environmental NGO in Denmark, held its
annual assembly. A resolution was adopted urging the Minister of the Environment, counties and
municipalities to substantially reduce loads from waste water treatments plants immediately and
to lessen losses from other activities, e.g. agriculture (ATV 1990). Almost in parallel, the Minis-
ter of the Environment developed and launched an “Action Plan for the Marine Environment”
(APME) for consideration and eventual adoption by the Parliament (Folketinget 1990). This
APME should be seen as a proposal from a Government without a majority in the Parliament —
hence, the plan was subject to political negotiations and eventual adoption.
On 18 November 1986, a majority in the Parliament forced the minority government, by adopt-
ing an official Parliamentary Agenda, to: (1) guarantee that all illegal discharges from municipal
waste water treatment plants, industries and agriculture would be brought to an end before 1 May
1987, and (2) to issue a Governmental Action Plan, including a plan for investments, aiming at a
reduction of N and P discharges with 50% and 80%, respectively, to be presented before 1 Feb-
ruary 1987 (Folketinget 1986-1987).
The Governmental Action Plan, published as “Action Plan Against Pollution of the Danish
Aquatic Environment with Nutrients” (Miljøministeriet 1987), was based on the Action Plan on
the Marine Environment taking the November agenda into account and was amended by the
Danish Parliament. Hence, no publication exists. The closest we come to a first Danish Action
Plan on the Aquatic Environment (APAE 1) reference is the combination of the Governmental
Action Plan (Miljøministeriet 1987) and the summary of changes adopted by the Parliament
(Folketinget 1986-1987). On the basis of a sequence of political agreements adopted by the Par-
liament’s Environmental and Planning Committee in April 1987, the Government pushed for-
ward a legislative process leading to adoption of a series of laws, including funding of monitor-
ing (1988-1991 and onwards) and research. In addition a suite of derived statutory orders imple-
menting specific elements of the APAE 1 were published. An overview can be found in ATV
(1990) and Miljøstyrelsen (1990).
20
The publication of the first nation-wide assessment of the state of the aquatic environment in
1990 (Miljøstyrelsen 1990) marks the end of the first cycle of the APAE. Hence, APAE 1 in-
cludes all four phases of AM: (1) Plan, (2) Act, (3) Check, and (4) Evaluate, collectively setting
the PACE not only for APAE 1, but also for the subsequent action plans.
The planning phase of APAE 1 focused strongly on the overarching aim of the APAE 1: 50%
reduction compared to the levels in the mid-80’s of nitrogen discharges and losses from agricul-
ture, urban waste water treatment plants (UWWTP), and industries with separate discharge. For
phosphorus the aim was to reduce discharges by 80%. Diffuse losses of phosphorus from agri-
culture were not included in APAE 1 owing to inaccurate data on this source. The Danish Par-
liament agreed on reduction targets (RT), reduction percentages (%) and target loads (TL), cf.
Table 2. It should remembered that the Action Plan from 1987, despite well-known uncertain-
ties, was based on an estimated annual loss of nitrogen from agriculture in the order of 260,000
tonnes TN. The reduction target (RT) was set to 127,000 tonnes corresponding to a reduction
percentage of 49%. Consequently, the resulting target load (TL) was set to be 133,000 tonnes.
Table 2: Danish nutrient reduction targets sensu the Action Plans for the Aquatic Environment 1 and 2. Baseline is 1987; reductions and targets were agreed by the Danish Parliament in 1987 and subsequently adjusted in 1990 (for UWWTP’s) and1999 (for industries). Units = tonnes per year. See Ærtebjerg et al. (2003) and Carstensen et al. (2006) for details.
Sector Total nitrogen loads (TN) Total phosphorus loads (TP)
1987 ÷ RT % = TL 1987 ÷ RT % = TL
Agriculture* 260,000 ÷ 127,000 49 = 133,000 4,400 ÷ 4,000 91 = 400 UWWTPs 18,000 ÷ 11,400 63 = 6,600 4,470 ÷ 3,250 73 = 1,220 Industries 5,000 ÷ 3,000 60 = 2,000 1,250 ÷ 1,050 84 = 200
Total 283,000 ÷ 141,400 50 = 141,600 10,120 ÷ 8,300 82 = 1,820
UWWTPs: Urban wastewater treatment plant effluents. RT: Reduction target. TL: Target loads. * Agricul-tural loads of phosphorus only concerns direct discharges from farms; diffuse losses of phosphorous were not included.
The act phase of APAE 1 focused on implementation of (1) measures to reduce nutrient dis-
charges, losses and emissions, (2) a nation-wide aquatic monitoring programme as well as (3)
two research programmes. It is beyond the scope of this thesis to go into details of the full pack-
age of laws and statutory orders enacted under APAE1. However, one piece of legislation – the
so-called ‘aktstykke’2 - is of particular interest since it laid down the economic basis of the Dan-
ish Aquatic Monitoring Programme (DAMP). DAMP was one of the most comprehensive na-
2 ’Aktstykke’ translates to a legal document, agreed by a majority in the Parliament, which secures sustained fund-
ing until the Parliament decides to terminate the funding.
21
tional aquatic monitoring programmes designed and carried out, with an action phase focusing
on (1) inclusion of all relevant sources (point and diffuse, the latter including atmospheric depo-
sition) and compartments (groundwater, lakes, streams and rivers, as well as marine water), (2)
coordination and documentation of all strategies and methods, (3) regional reporting, nation-
wide thematic assessments, and nation-wide integrated assessment, (4) regular evaluations and
revisions and (5) securing of funding (Folketinget 1986-1987, Indenrigsministeriet 1988,
Miljøstyrelsen 1989, Kronvang et al. 1993, Conley et al. 2002).
Two research programmes were initiated as part of the act phase, one on urban waste water
1987-1992 and another on marine eutrophication 1990-1994 (for details, see PH-Consult ApS
(1993), Jørgensen & Richardsson (1996), Christensen et al. (1998)). Aside from the underesti-
mated feature of sustained funding, the DAMP is unique because the periodic cycle of design,
sampling, evaluation and revision of monitoring activities. The monitoring programme that came
out of the planning and actions of APAE 1 (Miljøstyrelsen 1989 and Kronvang 1993) has result-
ed in four follow-up programmes: (1) DAMP 1993-1997, NOVA 1998-2003, NOVANA 2004-
2009, the latter including a so-called “half-way tuning” effected from 2007, and NOVANA
2010-2105 (see Miljøstyrelsen 1993, Miljøstyrelsen 2000 and Svendsen et al. 2004 for details).
The collaboration between the local partners (the counties) and the national partners (e.g. the
National Environmental Research Institute) has resulted in an accumulation of knowledge lead-
ing to many scientific publications, e.g. Conley et al. (2002), Kronvang et al. (2005), Carstensen
et al. (2006) and Kronvang et al. (2008). In addition to national assessment reports, the pro-
gramme has provided data for regional marine conventions such as HELCOM (e.g. HELCOM
2009) and OSPAR (e.g. Ærtebjerg et al. 2003 and OSPAR 2008).
The direct link from monitoring and assessment activities to evaluation of the action plan(s) was
already incorporated in APAE 1. The first evaluation of APAE 1 revealed that the reduction tar-
gets for discharges of nitrogen and phosphorus from urban waste water treatment plants and in-
dustries were likely to be met in the mid-1990s. The targets for discharges of both nitrogen and
phosphorus from urban waste water treatment plants were met in 1996 and 1995, respectively,
and are today below the targets of APAE 1 (Figure 3). For industries with separate discharge,
the targets for both nitrogen and phosphorus were met in 1996 and 1995, respectively, and are
today far below the targets (Figure 4).
22
A
0
5
10
15
20
Mid
80's
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
Nitrogen, 1000 tonnes TN
B
0
2
3
5
6
Mid
80's
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
Phosphorus, 1000 tonnes TP
Figure 3: Panel A: Temporal trend from the mid 1980’s to 2008 in nitrogen discharges from Danish urban waste water treatment plants to surface waters. Panel B: Temporal trend from the mid 1980’s to 2008 in
phosphorus discharges from Danish urban waste water treatment plants to surface waters. Data courtesy of the Danish Nature Agency – see Nordemann Jensen et al. (2010) and Annex A2.1 for details.
A
0
1
2
3
4
5
6
7
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
07
20
08
Nit
rog
en
, 1000 t
on
nes T
N
B
0
200
400
600
800
1000
1200
1400
1600
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
07
20
08
Ph
osp
ho
rus,
ton
nes T
P
Figure 4: Panel A: Temporal trends 1989-2008 in nitrogen discharges (tonnes of TN) from Danish indus-tries with separate discharge to surface waters. Panel B: Temporal trends 1989-2008 in phosphorus dis-
charges (tonnes of TP) from Danish industries with separate discharge to surface waters. Data courtesy of the Danish Nature Agency – see Nordemann Jensen et al. (2010) and Annex A2.2 for details.
The discharges of nitrogen and phosphorus from urban waste water treatments plants and indus-
tries with separate discharge are estimated by a specific sub-programme of DAMP (see page 18-
19). All industries are monitored in the same way and the estimated discharges are considered
accurate. However, it is noteworthy that data is reported with uncertainty. This might be a negli-
gable deficiency, since the contribution from industries with separate discharge is small. For ur-
ban waste water treatment plans, the loads are estimated by a complex combination of measure-
23
ments and statistical models (depending on size and type of UWWTPs). Hence, uncertainties
could be considerable and it would seem well-justified to analyse potential uncertainties.
The recognition of success of the plan in relation to point sources has lead to a very strong and
well-justified focus on the dominant source of nitrogen, the diffuse losses from agriculture. In
total, there have been three follow-ups on APAE 1:
1. The 1991 Action Plan of Sustainable Agricultural Development (APAE 1½).
2. The 1998 APAE 2, as a combined consequence of feedback from the DAMP activities, im-
plementation of the Nitrates Directive (Anon. 1991b) and a momentous oxygen depletion
event in Mariager Fjord (Fallesen et al. 2000).
3. The 2004 APAE 3.
Estimating losses from agriculture is more difficult to determine then it is for the discharges from
point sources. However, the nitrogen losses from agriculture can be calculated from the overall
nitrogen balance from agriculture, especially the estimate of the nitrogen surplus (Figure 5).
0
200
400
600
800
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
Nit
rog
en
, 1000 t
on
nes
N fertilizer Animal fertiliser Sludge N-fixation Deposition Harvest
Figure 5: Nitrogen input and export (harvest) in Danish agriculture 1985-2008. For details in regard to the nitrogen surplus, please see Nordemann Jensen et al. (2010) Annex A2.3.
Despite the clear trend in nitrogen surplus, it should be remembered that the APAE nitrogen tar-
get for agriculture was related to losses from fields. These losses from the root zone have been
monitored and assessed by a specific sub-programme of DAMP. Here, the strategy has been to
intensively monitor six small-scale agricultural catchments. The trends in nitrogen losses from
the root zone in monitored agricultural catchments are outlined in Figure 6.
24
0
20
40
60
80
100
120
140
160
180
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
N lo
ss
(k
g N
/ha
)
LOOP6
LOOP2
LOOP3
LOOP4
LOOP7
LOOP1
Figure 6: Estimated losses of nitrogen (kg/ha) in six Danish agricultural catchments (LOOP areas) 1991-
2008. From Nordemann Jensen et al. (2010). The weighted reduction is presented in Annex A2.4.
The selection of the agricultural catchments (LOOP areas) and the monitoring activities are well-
justified. However, the sub-programme cannot be fully representative given the variation in soil
types and agricultural activities in Denmark. It is difficult to grasp why a programme taking spa-
tial and temporal variations into account has not been developed. An answer, which is specula-
tive, is that the sub-programme is unpopular with stakeholders living within the catchments and
politically sensitive. Further, the sub-programme is faced with another challenge. The up-scaling
from relatively few catchments using simple statistical models, to a nation-wide estimate leaves
considerable room for improvement. No assessments of uncertainties in the annually estimated
losses have been published throughout the APAE period. Considering that these estimates are the
basis for evaluation of whether Danish agriculture has met the targets for reduction of nitrogen
discharges according to the APAE’s, it is beyond understanding that obvious ways to reduce
uncertainties, e.g. by setting up a spatially representative sampling programme and by using
complex models, have apparently been ignored.
It would be prudent to mention that uncertainties in regard to the nitrogen losses were dealt with
in connection with the evaluation of APAE 2 and setting up of APAE 3. It was documented that
the original estimated losses of nitrogen from cultivated fields were underestimated (Grant &
Waagepetersen 2003). Hence, the loss was corrected to 311.000 tonnes of TN, or 20% higher
compared to APAE1. Table 3 is based on Table 2 but updated according to APAE 3.
25
Table 3: Danish nutrient reduction targets sensu the Action Plans for the Aquatic Environment 1, 2 and adjusted sensu Action Plan 3. From Andersen & Conley (2009), based on Carstensen et al. (2006).
Sector Total nitrogen loads (TN) Total phosphorus loads (TP)
1987 ÷ RT % = TL 1987 ÷ RT % = TL
Agriculture* 311,000 ÷ 152,400 49 = 158,600 4,400 ÷ 4,000 91 = 400 UWWTPs 18,000 ÷ 11,400 63 = 6,600 4,470 ÷ 3,250 73 = 1,220 Industries 5,000 ÷ 3,000 60 = 2,000 1,250 ÷ 1,050 84 = 200
Total 334,000 ÷ 166,800 50 = 167,200 10,120 ÷ 8,300 82 = 1,820
Hence, a correction factor of 1.2 can be used for ‘normalising’ APAE 1, the Action Plan for Sus-
tainable Agricultural Development and APAE 2 with APAE 3 – such correction enabled direct
comparisons between APAE 3 and its predecessors and can be seen in Table 4. In the columns
labelled “old”, the figures are based on APAE 1, while the columns labelled “new” includes fig-
ures corrected by a factor of 1.2. The revised values are 127.000 or 152.400 tonnes, cf. Tables 2
and 3. Please note that the evaluation of APAE 1 is shown under APAE 1½ as “old”, the evalua-
tion of APAE 1½ is shown under APAE 2 as “old” and the evaluation of APAE 2 is shown un-
der APAE 3 as “old”.
Table 4: Summary of planned and implemented reduction targets in regard to nitrogen discharges and losses from agriculture according to the Danish Action Plan 1 from 1987, the subsequent Action Plan for Sustainable Agricultural Development from 1991 (here named APAE 1½) and the follow up Action Plans 2 and 3 from 1998 and 2004. Unit = tonnes TN per year. See Ærtebjerg et al. (2003) for evaluation re-sults (red numbers) as well as details about the specific measures under the Action Plans 1, 1½ and 2.
APAE 1 APAE 1½ APAE 2 APAE 3
Old New Old New Old New New
Action Plan 1 127,000 152,400 50,000 60,000 — — — —
Action Plan 1½ — — 77,000 92,400 89,900 107,880 — —
Action Plan 2 — — — — 37,100 44,520 113,460
Action Plan 3 — — — — — — 38,940
Grand total 127,000 152,400 127,000 152,400 127,000 152,400 152,400
The first evaluation phase of the APAE 1 as well as the evaluation of subsequent action plans
provides excellent examples of adaptive management. This is evident in the way information
from the monitoring programme was analysed, compared to the goals of APAE 1, synthesised
and eventually directed to the Minister of the Environment for political considerations and sub-
sequent tightening of the measures needed. The last outcome of this evidence-based AM loop
included in the thesis is the 2008 mid-term evaluation of APAE 3 (Waagepetersen et al. 2008).
As word of caution it should be noted that APAE 2 introduced a measure related to reconstruc-
tion of wetlands. Consequently, the evaluation criteria not only include losses from the root zone,
26
but also retention in wetlands and streams. Whilst the estimation of losses from the root zone is
made by DAMP, the estimation of the retention in wetlands and streams has a more uncertain
origin. No nation-wide technical guidance is available, instead a variety of estimates ranging
from the use of empirical models to complex hydrological and biogeochemical models are used.
The 50% reduction target for nitrogen was estimated to be met via the full implementation of
APAE 3 (Waagepetersen et al. 2008). The legitimacy of this estimate is yet to be determined as it
is based on the monitoring of agricultural watersheds, which apparently is a non-representative
data set. However, the estimate is indirectly confirmed by sub-program of DAMP dealing with
inputs to marine waters (Nordemann Jensen et al. 2010, Carstensen et al. 2006).
1.2: The Baltic Sea Action Plan
The countries surrounding the Baltic Sea have since 1974 joined forces in order to safeguard the
Baltic Sea environment and to coordinate mitigatory efforts. The framework for this work is the
‘Convention on the Protection of the Marine Environment of the Baltic Sea Area’ – known as the
Helsinki Convention. The governing body is the Helsinki Commission, which is responsible for
the coordination of activities and day-to-day work.
Nutrient enrichment and eutrophication were dealt with for the first time at a high political level
by the Ministers of the Environment of the Baltic Sea States at a Ministerial Meeting in 1988.
The first Danish Action Plan on the Aquatic Environment played a key role in regard to this Bal-
tic Sea-wide adoption of the 50% reduction target, which were stated in the 1988 Ministerial
Declaration:‘…efforts on reduction of the load of pollutants should aim at a substantive reduc-
tion of the substances most harmful to the ecosystem of the Baltic Sea, especially of heavy metals
and toxic and persistent organic substances, and nutrients for example in the order of 50 percent
of the total discharges of each of them, as soon as possible, but not later than 1995’ (HELCOM
1988).
The Helsinki Convention was revised in 1992 in order to embrace the changed geopolitical situa-
tion. The new convention also became more explicit in regard to eutrophication, e.g. by includ-
ing an annex with a specific focus on the needs for reducing water-borne nutrient inputs from
point and diffuses sources. In order to support the implementation of the reduction targets agreed
upon at the 1988 Ministerial Meeting, the Baltic Sea Joint Comprehensive Environmental Action
Programme (JCP) was established in 1992. Identification and elimination of pollution Hot Spots
27
was an important part of this work, and initially 132 Hot Spots were identified including both
municipal wastewater treatment plants and agricultural ‘sites’. In 2002, an evaluation of
achievements revealed that the 50% reduction target for the time period from 1987 to 1995 had
been achieved for phosphorus discharges from point sources by almost all countries, while most
countries had not reached the targets for nitrogen (Lääne et al. 2002). Agricultural loading levels
showed smaller decreases than point-source loading despite the fact that almost all countries in
transition3 had achieved the 50% target for phosphorus. However, accurate estimates of changes
in agricultural loading were hampered by a lack of monitoring data. Further estimation of
achievements between 1985 and 2000 showed that as a result of improved treatment of industrial
and municipal wastewaters, nutrient discharges from point sources had greatly decreased.
The reduction targets for diffuse sources such as agriculture were not fulfilled (HELCOM 2009).
Hence, it remained clear that eutrophication was still of concern. The HELCOM Bremen Minis-
terial Meeting Declaration of 2003 demanded further actions, in particular in the agricultural
sector, to reduce diffuse nutrient loads. In addition, HELCOM was tasked to implement an eco-
system approach to the management of human activities and the idea of developing ecological
objectives with indicators was put forward.
In 2006 HELCOM adopted a system of ecological objectives with the specific strategic goal for
eutrophication of a ‘Baltic Sea unaffected by eutrophication’ defined by five specific ecological
objectives: (1) ‘concentrations of nutrients close to natural levels’, (2) ‘clear water’, (3) ‘natural
levels of algal blooms’, (4) ‘natural distribution and occurrence of plants and animals’, and (5)
‘natural oxygen levels’. To make these ecological objectives operational, indicators with initial
target values were agreed upon reflecting a good ecological and environmental status of the Bal-
tic marine environment. Thus, the target values, when achieved, are intended to represent good
ecological or environmental status. It has subsequently been agreed that the ecological objectives
for eutrophication will be measured by the following indicators: (1) winter surface concentra-
tions of nutrients, reflecting the ecological objective ‘concentrations of nutrients close to natural
levels’; (2) Chlorophyll-a concentrations, reflecting the ecological objective ‘natural level of
algal blooms’, (3) Secchi depth, reflecting the ecological objective ‘clear water’, (4) depth range
of submerged aquatic vegetation, reflecting the ecological objective ‘natural distribution and
occurrence of plants and animals’, (5) abundance and structure of benthic invertebrate communi-
3 Estonia, Latvia, Lithuania and Poland.
28
ties, reflecting the ecological objective ‘natural distribution and occurrence of plants and ani-
mals’, and (6) area and length of seasonal oxygen depletion, reflecting the ecological objective
‘natural oxygen levels’. More information in regard to the operationalization of the above indica-
tors can be found in Andersen et al. (2004, 2006), HELCOM (2006), and HELCOM (2009).
To have a more targeted approach to address the symptoms of eutrophication, it was considered
necessary to have nutrient reduction targets taking into account both ecosystem functioning and
sub-regional differences. A model-based approach employing sub-regional targets related to se-
lected ecosystem features such as water transparency was established (Wulff et al. 2007). Fol-
lowing the principle of adaptive management and in order to implement the ecosystem approach
to the management of human activities, HELCOM coordinated the development of the Baltic
Sea Action Plan (BSAP).
Initial estimates of nutrient reductions needed to reach the target levels for eutrophication were
produced by the MARE program (Wulff et al. 2007). In addition, scenarios were considered to
examine how far the full implementation of existing HELCOM Recommendations, as well as
EU legislation and programmes, would bring the Baltic Sea towards the agreed ecological objec-
tives for eutrophication, using the target ‘clear water’ as a basis. These results produced by
MARE were used to develop specific reduction targets and actions related to reducing nutrient
loading to the BSAP. Hence, the BSAP defines maximum nutrient loads that will allow
achievement of eutrophication targets for the whole Baltic Sea and each of its sub-basins. The
required reductions in nutrient loads were estimated based on the objective ‘clear water’, model-
ling of maximum allowable nutrient loads matching the objective, and average nutrient load lev-
els from 1997 to 2003. It was acknowledged that the maximum allowable nutrient loads and the
country-wise allocations of the BSAP were based on the best knowledge at the time and that
review and revision of the figures should start as soon as the BSAP was adopted. By using an
evidence-based target, the BSAP is partly, but not completely based on the Ecosystem Approach.
The BSAP contains measures estimated to be sufficient to reduce eutrophication to a target level
that would correspond to good ecological/environmental status by the year 2021 (HELCOM
2007). It was estimated that nutrient load reductions of 135,000 t of nitrogen and 15,250 t of
phosphorus from average annual nutrient loads (based on loads during the period 1997–2003)
would be needed. Quantitative reduction requirements were applied to each of the sub-basins and
29
provisional allocations of nutrient reduction requirements to each HELCOM country and to
transboundary loads were included in the BSAP. The main bulk of reductions were to be made in
the Baltic Proper, while the Gulf of Bothnia was at that time considered to have good ecologi-
cal/environmental status and thus not in need of reductions. It was estimated that the reductions
would result in achieving the eutrophication-related targets on water transparency, primary pro-
duction and nutrient concentrations (Wulff et al. 2007). The inputs to and outputs from the
MARE/NEST calculations on maximum allowable loads are summarised in Table 5. Time de-
lays in achieving good ecological status were presumed to be significant, on the order of decades
due to long residence times, even in the case that all nutrient reductions were made immediately
(Savchuck & Wulff 2007).
Table 5: Maximum allowable annual loads of phosphorus and nitrogen to achieve ‘good environmental status’ (calculated for water transparency) and corresponding minimum load reductions (in tonnes) calcu-lated per sub-basin (based on HELCOM 2007).
Basin Maximum allowable nutrient loads (tonnes)
Inputs in 1997–2003 (normalized)
Needed reductions (interim allocation)
TP TN TP TN TP TN
Bothnian Bay 2,580 51,440 2,580 51,440 0 0 Bothnian Sea 2,460 56,790 2,460 56,790 0 0 Gulf of Finland 4,860 106,680 6,860 112,680 2,000 6,000 Baltic Proper 6,750 233,250 19,250 327,260 12,500 94,000 Gulf of Riga 1,430 78,400 2,180 78,400 750 0 Danish Straits 1,410 30,890 1,410 45,890 0 15,000 Kattegat 1,570 44,260 1,570 64,260 0 20,000
Sum 21,060 601,710 36,310 736,720 15,250 135,000
The country-wise allocation of re-
ductions is summarized in Table 6.
In order to reduce nutrient inputs to
the Baltic Sea to the maximum al-
lowable level the countries have also
agreed to take actions not later than
2016 to reduce the nutrient load
from waterborne and airborne inputs
aiming at reaching good ecological
and environmental status by 2021.
Table 6: Country-wise nutrient load reduction allocations, in tonnes (from HELCOM 2007).
Country Reductions
TP TN
Denmark 16 17,210 Estonia 220 900 Finland 150 1,200 Germany 240 5,620 Latvia 300 2,560 Lithuania 880 11,750 Poland 8,760 62,400 Russia 2,500 6,970 Sweden 290 20,780 Transboundary pool 1,660 3,780
Total 15,016 133,170
30
Figure 7: Nitrogen and phosphorus loads to the Baltic Sea 1990-2006. The target loads of the BSAP are indicated as ‘2021’. From Andersen et al. (2010).
Meeting the 2007 BSAP targets by 2021 will be a difficult and would be a significant achieve-
ment. However, it should be recognised that the Baltic Sea States have already reduced input of
nutrients significantly, especially for phosphorus (Figure 7).
The BSAP does not include a well
described checking phase (Figure 8).
However, this is not an issue since
monitoring is already dealt with via the
HELCOM monitoring and assessment
strategy (HELCOM 2005) and the
HELCOM COMBINE programme
(HELCOM 2008). Assessments in-
clude annually updated HELCOM
Indicator Fact Sheets and the produc-
tion of thematic assessment reports on
eutrophication (e.g. HELCOM, 2009),
which covers the period 2001-2006
and sets a baseline for the BSAP. Further, the BSAP does not include an unambiguous evalua-
tion phase. This may not be a significant issue, since the countries have committed themselves
politically to (1) implement AM for the restoration of good ecological/environmental status of
the Baltic Sea, and (2) revisiting the nutrient reduction targets and measures, in particular the
country-wise allocation. The upcoming 2013 HELCOM Ministerial Meeting will evaluate the
effectiveness of the national programmes and review the progress towards the ecological objec-
tives describing a Baltic Sea in good status.
Figure 8: Illustration of the core of the eutrophication segment in the BSAP. The first step is political, the
second step is in principle scientific (setting target is the basis of the politically agreed visions and objec-tives) while the third step is a combination of science (scenario modelling) and policy (agreeing on the reduction targets). Based on HELCOM (2007) and Backer (2008).
31
2: Similarities and Differences Between the Danish
Action Plans on the Aquatic Environment and
the Baltic Sea Action Plan
The origin of the first Danish Action Plan on the Aquatic Environment (APAE 1) and its evolu-
tion are summarized in Figure 9. Many direct and indirect connectors are identified, the most
important ones are highlighted in the following sections.
The Belt Project and especially its successor, the National Pollution Monitoring Programme
identified a large-scale eutrophication problem in Danish marine waters. This, in combination
with results originating from regional monitoring (including ground and freshwaters), led to a
good understanding and wide acceptance of the cause-effect relations leading to oxygen deple-
tion in the inner Danish marine waters.
The interactions between research and monitoring and subsequent evaluations of the APAE’s
have proved to be working as intended by APAE 1. Of particular importance are the links from
the Danish Marine Environmental Research Programme, initiated via APAE 1, to the revisions
of national monitoring programs leading to the Danish Aquatic Monitoring Program (DAMP)
1993-1997 and NOVA-2003. Three elements of the monitoring programs worth highlighting are:
(1) weekly sampling at open water stations in dynamic areas, (2) coastal areas with extended
sampling programmes including mass balances for nutrients, and (3) the inclusion of a 3D Ma-
rine Modelling Complex covering open Danish marine waters, including the North Sea, Skager-
rak, Inner Danish Waters, and the western Baltic Sea.
Input from the nation-wide monitoring programmes (DAMPs, NOVA, NOVANA) have been
directly linked to revisions of the APAEs: (1) The Action Plan for Sustainable Agricultural De-
velopment (from 1991, sometimes called APAE 1½ ) was very much influenced by DAMP
1989-1992, (2) APAE 2 was directly influenced by DAMP 1993-1997 as well as implementation
of the Nitrates Directive, the latter being amplified by the incident in Mariager Fjord where the
whole water column became anoxic, and (3) APAE 3 was influenced by NOVA-2003, in par-
ticular the sub-programmes for catchment monitoring and riverine loads. Hence, the monitoring
programs have been and still are a backbone, providing data for assessments and evaluations.
32
Figure 9: Key interactions between the Danish Action Plans on the Aquatic Environment, monitoring of
the aquatic environment and marine research. Colours refer to a pre-phase (white), the Action Plans as such and derived activities (grey) and indirectly related activities (light grey).
An overlooked and to some extent refreshing feature in regard to the Danish APAE 1 is that it is
evidence-based, both in regard to its roots (being the Belt Project and its successor, the National
Pollution Monitoring Programme (Ærtebjerg Nielsen et al. 1981, ATV 1990)) and partly also in
regard to its reduction targets (Miljøstyrelsen 1987 &1990 and Jens Brøgger, pers. comm.). The
APAE 1 has from time to time been criticized for being based on uncertain estimates of the exist-
ing loads, especially the losses from agriculture, and the required reductions (ATV 1990). Parts
of the criticism are understandable, but perhaps not entirely justified. The Government’s Action
Plan for the Marine Environment (APME), developed and published by Miljøstyrelsen (1987),
was evidently not based on a nation-wide mass balance or comprehensive scenario modelling,
33
but still it was based upon the best available information (e.g. the NPo Action Plan (Folketinget
1984-85)) as well as justified estimates for load reduction prepared by the counties (e.g.
Hovedstadsrådet 1983). Given the media pressure at that time, the criticism seems to underesti-
mate an inevitable political need for action.
There is no single element in the Danish APAE’s that makes them successful; it is a combination
of activities. Four elements are likely to be of particular significance. First of all, it is the con-
sistent, direct and sustained implementation of measures, especially in regard to discharges of
phosphorous from point sources and losses of nitrogen from diffuse sources. Secondly, it is the
monitoring. Thirdly, it is both the regional and national assessments of environmental status.
Fourthly, it is a solid political will, at least in the first decade of the APAE’s, to follow up on the
results of the evaluations, especially when it comes to APAE 1 and 2.
A key characteristic of the Danish APAEs is their focus on all four phases of AM, e.g. the PACE
principles. APAE 1 in particular had a strong focus on Planning (in particular the overall reduc-
tion targets, well designed monitoring and assessment systems, the addition of a pre-planned
evaluation), on Actions (in particular the focus on relevant sources and reduction of both N and
P), on checking (in particular establishment of DAMP 1989-1992, annual reporting at three lev-
els, sustained funding) and on a pre-planned Evaluation (in particular the first evaluation and the
agreement on the Action Plan on Sustainable Agricultural Development (APSAD being nick-
named APAE 1½), which in reality primed the basis for all subsequent evaluations and follow up
plans).
Another characteristic is the role of science: the Belt Project and the NPo Research Programme
initially fed results into the APs and monitoring programmes. Especially the Danish Marine En-
vironmental Research Programme (Jørgensen & Richardson (1996), Christensen et al. (1998))
played a crucial role, not only in regard to re-design of the marine monitoring programmes but
also in regard to building a joint and widely accepted understanding of causes and effects of eu-
trophication as well as capacity building locally (counties/environmental centres) and nationally
(in particularly at the National Environmental Research Institute). In later years, another im-
portant feature has been the vast number of scientific publications based on the long-term moni-
toring of inputs and effects in all compartments of the aquatic environment (e.g. Conley et al.
34
2000, Nielsen et al 2002a, Nielsen et al 2002b, Ærtebjerg et al. 2003, Josefson & Hansen 2004,
Carstensen et al. 2006, Conley et al. 2007, and Carstensen et al. 2011).
When scrutinising what has happened during the evaluation of the Danish Action Plans, not eve-
rything is clear-cut. When the 1987 reference losses for root zone nitrogen from agriculture were
corrected in 2003, as preparation for the 2004 AP III, they were assumed to be 20% higher than
earlier estimated (260.000 to 311.000 t cf. Table 3) (Grant & Waagepetersen 2003). However,
the percentage used for the correction of the 1987 reference loss could be a negotiated figure and
there was no scientific, public or political debate of the corrected figure and in particular of its
implications.
Some points worth reiterating are that the APAE 1 overruled the different agendas adopted by
the Parliament and that the APAE had three ‘targets’: (1) a reduction target (RT), a reduction
percentage (%) as well as a target load (TL). While APAE 1 and 2 focused on the numbers in the
original AP (See Table 3), the APAE 2 evaluation, because of the correction of the reference
load for nitrogen from agriculture (see Table 4), created an opportunity for APAE 3 to put em-
phasis on either the reduction target or the reduction percentage or the target load, cf. the scenar-
ios presented in Table 7. An interpretation of the three scenarios presented in Table 7 could be
that the course set by APAE 3 is likely to be the least stringent. The scenarios being more strin-
gent in regard to alleviation of eutrophication were simply disregarded by APAE 3. A likely ex-
planation would be that the agricultural sector had regained a political influence as strong as be-
fore the adoption of APAE 1.
Table 7: Summary of estimated differences between the 1987 load targets in Action Plan I. Scenarios are based on the corrected reference loads and setting of a fixed reduction target (scenario I), a fixed reduc-tion percentage (scenario II), and a fixed load target (scenario III). Numbers in red originates from APAE 1. The difference is calculated as the 1987 load target minus the revised load target. Units = tonnes TN.
AP 1 Scenario I Scenario II Scenario III
Reference loads from agriculture 260,000 311,000 311,000 311,000 Reduction target 127,000 127,000 152,400 178,000 Reduction percentage (%) 49% 41% 49% 57%
Load target 133,000 184,000 158,600 133,000
Difference - ÷51,000 ÷25,600 0
Another remarkable point is that the APAE 1 seems to be inspiration for other European 50%
reduction plans, e.g. those by HELCOM and OSPAR, which were agreed in February 1988 and
in June 1988, respectively (OSPAR 1988, HELCOM 1988). Denmark promoted key principles
35
originating from the APAE 1, but it is unclear if the countries around the Baltic Sea and the
North Sea simply acknowledged the Danish adoption of the APAE 1 or agreed on the necessity
of almost identical reduction targets.
The overture leading to the adoption of the HELCOM Baltic Sea Action Plan (BSAP) can be
described as protracted. However, it reflects the scientific understanding of eutrophication in the
Baltic Sea (Elmgren 2001) as well as the shift in geopolitical conditions. The BSAP can be said
to have had a long prelude, starting with the 1974 Convention and the 50% reduction target in
the 1988 Ministerial Declaration. Reductions in nutrient loadings have been achieved by most
Baltic Sea countries; the long-term results are remarkable while the recent short-term develop-
ments (2004–2006) are not as encouraging. The reductions have not yet resulted in a Baltic Sea
unaffected by eutrophication. Hence, the good environmental status in terms of eutrophication as
defined by the HELCOM BSAP had not been reached via the predecessors to the BSAP. The
links between the 1974 Convention and the onset of regular Baltic Sea-wide assessments includ-
ing assessments of eutrophication status are illustrated in Figure 10.
The drivers that will result in a decrease in loads are proper implementation of national action
plans and HELCOM recommendations as well as a number of legally binding international
agreements and legislation such as the European directives addressing eutrophication. The most
recent additions to the list of drivers are the BSAP and the Marine Strategy Framework Directive
(MSFD). Implementation of the Urban Waste Water Framework Directive (UWWTD), Nitrate
Directive (ND), Water Framework Directive (WFD) and MSFD is essential, because tangible
and durable improvements in the eutrophication status of the Baltic Sea rely on the load reduc-
tions provided via these directives and without their proper implementation, progress, if any, will
be very slow and difficult to document. Moreover, the implementation of these directives has al-
ready been taken into account when establishing the eutrophication segment and load reduction
allocations of the BSAP.
The eutrophication segment of the BSAP envisages provisional national load reductions tenta-
tively set up on the basis of: (1) overall objectives and a set of targets for water transparency, (2)
model calculations of maximum allowable loads and country-wise reduction targets, and (3) re-
duction scenarios and cost-efficiency. The approach employed is well-justified and well-
documented and should be seen as an appropriate first step. The BSAP thus acknowledges that
36
the figures related to targets and maximum allowable nutrient loads should be periodically re-
viewed and revised using a harmonized approach based on the most recent information and data.
Figure 10: Key interactions between political agreement under the Helsinki Conventions (Conventions,
Ministerial Declarations and Action Plans), Baltic Sea-wide assessment of the state of the environment and other eutrophication related policy drivers. BIO = integrated thematic assessment of biodiversity and nature conservation; BSAP = Baltic Sea Action Plan; EUT = integrated thematic assessment of eutrophi-cation; HAZ = integrated thematic assessment of hazardous substances; HOLAS = Holistic Assessment 2003-2007; JCP = Baltic Sea Joint Comprehensive Environmental Action programme; MSFD = Marine Strategy Framework Directive; WFD = Water Framework Directive; UWWTD = Urban Waste Water Treatment Directive; ND = Nitrates Directive.
37
Hence, the BSAP includes an element of Adaptive Management, but no progress reporting or
evaluations as such. Further technical development of the modelling approach should be carried
out by including a broader range of indicators, such as nutrient concentrations, chlorophyll-a
concentrations and oxygen in addition to the currently employed water transparency. In addition,
greater coherence is needed between the modelling approach and the practical use of modelling
results, and most likely also future eutrophication assessments. Coherence could be enhanced by
increasing the temporal resolution of the model to the level which is employed, inter alia, in this
status assessment enabling a distinction between the different seasons instead of data averaged
over the annual cycle. This would not only improve the reliability of the approach and load allo-
cations, but also lead to greater credibility among the public, which has not yet been achieved by
any regional marine convention.
The total acceptable loads sensu the 2007 HELCOM BSAP and the 50% reduction targets sensu
the 1988 HELCOM Ministerial Declaration cannot directly be compared because there are dif-
ferences in the approaches used. An indirect comparison indicates that the 2007 BSAP is stricter
in terms of phosphorus than the 1988 Ministerial Declaration. In terms of nitrogen, however, it
could appear that the 1988 Ministerial Declaration might be stricter. Nonetheless, this may not
be significant for the following reasons: (1) the BSAP, addressing eutrophication using a holistic
ecosystem approach, specifies a number of indicators with associated targets which are compa-
rable with what would have been the ultimate effect of implementing the 50% reduction target,
(2) the BSAP does not (yet) take a consistent implementation of the WFD into account in terms
of expected load reductions, and (3) the BSAP will pursue declining loads and allow the Baltic
Sea to recover from its present status.
A characteristic of the BSAP eutrophication segment is that it builds on a suite of principles
which in combination makes it ecosystem based. The principles include, cf. Fig. 5: (1) setting of
visions, objective and selection of indicators (Backer and Leppänen (2008), (2) operational tar-
gets (EutroQOs) based on RefCon and AcDev (Andersen et al. (2004, 2006, 2010), HELCOM
(2009), Andersen (2010)), (3) linking targets and loads leading to estimation of Total Allowable
Loads and scenarios for cost-effective country-wise reduction targets (Wulff et al. 2007) and
subsequent, but not yet implemented (4) actions according to the Water Framework Directive,
which shall be fully implemented by 2016. However, when scrutinising the basis of the BSAP
eutrophication segment, it may not look as good as claimed. The BSAP is currently based on a
38
single indicator/target i.e. water transparency (Secchi depth). This does, at least in principle, base
the BSAP on Ecosystem Based Management (EBM). However, it might appear that the BSAP
has been rushed since it is not based on any other targets/indicators such as causative factors
(e.g. nutrient concentrations), primary effects (e.g. primary production, chlorophyll-a), or sec-
ondary effects (e.g. changes in benthic communities, oxygen concentrations). To be ecosystem-
based beyond doubt would require inclusion of more targets/indicators, cf. Figure 11.
Revision of the load calculations would also be problematic if not based on: (1) updated load
figures including atmospheric deposition, (2) an update of the targets and (3) inclusion of more
targets/indicators. HELCOM (2009) and Andersen et al. (2010, 2011) provide valuable infor-
mation in regard to target setting for all major basins of the Baltic Sea. Further development of
the model is in progress (Maria Laamanen, pers. comm.), and taking this into account when up-
dating the allowable loads and the load allocation would turn the eutrophication segment of the
BSAP into a state-of-the-art ecosystem-based nutrient management strategy.
Figure 11: Suggested framework for implementation of the eutrophication segment of the Baltic Sea Ac-
tion Plan. Please note that this framework is ecosystem-based, taking relevant eutrophication effects into account, e.g. elevated nutrient concentrations, primary effects (e.g. Chlorophyll-a concentrations) and secondary effects (e.g. benthic communities and oxygen depletion). Based on HELCOM (2009).
39
An important similarity between the Danish Action Plans and the Baltic Sea Action Plan is that
both are closely related to the implementation of a suite of eutrophication-related EU directives,
e.g. the EC Urban Waste Water Treatment Directive (Anon 1991a), the EC Nitrates Directive
(Anon 1991b), the EU Water Framework Directive (Anon 2000) as well as the EU Marine Strat-
egy Framework Directive (Anon 2008). The interactions between these eutrophication related
directives are many and complex (HELCOM 2009), but in no way conflicting. The directives are
despite differences in focus and terminology all striving toward a better ecological status of sur-
face waters and are focused on cuts in nutrient loads.
Both the Danish APAE’s and the BSAP are based on a political agreement but the Danish
APAE’s are enacted by national law and statutory order while the BSAP is implemented via ‘soft
law’ such as the Helsinki Convention and HELCOM Recommendations and not legally binding.
The Danish APAE can be viewed as “dark green” while the BSAP is “light green” in the sense
that the APAE are enacted whilst the BSAP is incentive based sensu Ernst (2010), who have
characterized the spectrum of “Mainstream Environmental Thought” as “dark green” (left wing
with a preference for mandatory agreements), “light green” (moderate with a preference for vol-
untary agreements) and “conservative” (right wing with a preference for market-oriented solu-
tions).
Another significant difference is the strong focus of the Danish APAE’s on progress reporting
based on monitoring and assessment and evaluations where the progress is compared to the goals
of the APAE 1. This is in line with the concept of Adaptive Management, where an Evaluation
phase is an integral and re-occurring element. Although the BSAP is in its early phases of im-
plementation, it could appear that there is a risk of losing momentum, e.g. in regard to: (1) the
recalculation of the load allocation, (2) progress reporting, and (3) evaluations where the pro-
gress is compared to the already agreed load reductions. The advantage of being based on the
concept of Ecosystem-Based Management might not be as great as assumed, and even out-
weighed by the weakness of not having included a well-planned evaluation phase with ‘rules of
procedure’ if loads are not reduced as agreed. Finally, another difference is that that Danish
APAE’s are focused on discharges and losses to freshwater and marine waters, while the BSAP
is focused only on inputs to the sea.
40
Based on the Danish Action Plans on the Aquatic Environment and the HELCOM Baltic Sea
Action Plan, four key characteristics of the plans have been identified:
1. A good nutrient management strategy (NMS) must include four well defined phases: (1)
Planning, (2) Action, (3) Checking, and (4) Evaluation. No weak links are allowed in this
PACE sequence. A perfect plan and accurately estimated reduction needs are of no value
without monitoring and regular evaluation because improvements in water treatment, agricul-
tural practises and ultimately eutrophication status span over timescales longer than political
election periods. High-quality monitoring without progressive reductions of loads will not re-
sult in improvement of environmental quality. The Danish APAE’s, being strong in all phas-
es, are currently the only NMS that successfully have proven to reduce inputs from both
point sources and diffuse sources (Carstensen et al. 2006, Kronvang et al. 2008).
2. A good NMS must have overarching goals and reduce loads of relevant nutrients, in particu-
lar nitrogen and phosphorus. Whether the goals are reduction goals as in the Danish APAE’s,
eutrophication targets (EutroQOs) or a combination of EutroOQs and reduction targets as in
the BSAP, is of minor importance. The key thing is that goals are being set and are easy to
understand. The reduction targets of the Danish APAE were clear and unambiguous, and the
BSAP objectives of total allowable loads and country-wise load allocation are in principle
not open to interpretation. And, as illustrated by the Danish APAE, some degree of uncer-
tainty in regard to the target setting does not affect the likelihood of fulfilling the plan as long
as loads are progressively reduced.
3. The effects of a good NMS must be documented by high-quality monitoring and assessment
activities. Here, the monitoring activities under the Danish APAE’s are examples for others
to learn from.
4. A good NMS relies upon political will to carry out evaluation as planned and, above all, to
conclude and follow up. Neither the best available scientific advice nor good planning can
substitute a sustained political will to alleviate eutrophication symptoms. Both the Danish
APAE’s and the BSAP are based on a political will to lessen eutrophication effects, and the
long-term sustained political will in Denmark, especially in the period 1986-2001, is perhaps
exceptional.
In addition to the key characteristics above, a number of specific requirements in regard to suc-
cessful implementation of the PACE sequence are identified:
1. Planning, meaning both agreeing on overall objectives and targets and developing a strategic
plan on how to achieve these should, on the basis of the analyses of the two case studies, in-
clude:
a) A document including not only political objectives and targets but also description of and
strategies for the three other phases (Actions, Checking and Evaluation), and
b) A plan for multi-year funding of actions (measures) and monitoring.
2. Actions, understood as agreeing on measures and subsequent implementation of these,
should on the basis of the two case studies analysed include:
a) Reduction targets for both nitrogen and phosphorus – and covering all contributing sec-
tors (e.g. agriculture, industries, households, energy production, and transport), and
b) Legal instruments such as laws, statutory orders (as in the Danish APAE’s) or alterna-
tively recommendations (as in the case of the BSAP).
41
3. Checking, understood as monitoring and assessment, should on the basis of the case studies
include:
a) Effects and inputs, a marine programme alone is not enough, information about inputs
and activities in upstream catchments is essential, cf. the Danish Aquatic Monitoring
Programme (DAMP), and
b) Adequate spatial and temporal coverage of primary and secondary effects, nutrients, in-
puts, sectors, human activities.
c) A detailed and unambiguous programme manual including description of methods, data
flow, and quality assurance/quality control procedures, and
d) Publication of assessments reports, in principle covering all relevant compartment of the
aquatic environment (point sources, diffuse sources, groundwater, lakes, rivers and ma-
rine waters) – in order to maintain awareness.
e) Evidently, sustained funding is an immense advantage, alternatively multi-year funding
e.g. for a 4-6 year period can be used as a guiding principle.
4. Evaluation, especially in the understanding of a strategic check as to whether the objectives
and target are being fulfilled, should include evaluation criteria as well as actions to be
agreed during the planning phase.
The Danish APAEs and the HELCOM BSAP share many principles including the PACE se-
quence. However, there are three major differences:
First, the BSAP is still in its first round, while the Danish APAE’s are approaching a fourth
round. Therefore, the interim judgement of the BSAP’s checking and evaluation phases
might turn out to work better than predicted.
Second, the Danish APAEs are adopted by a majority in the Danish Parliament and imple-
mented in national law. The HELCOM BSAP is not legally binding, it is merely a political
expression of interest. Since it is closely linked to a suite of EU directives, in particular the
Urban Waste Water Treatment Directive (UWWTD) and Marine Strategy Framework Di-
rective (MSFD), this might compensate for the non-legally binding character of the BSAP.
Third, whilst the APAEs are focusing reduction of discharges and losses from three sectors,
the BSAP is focusing on an ecological target being ‘clear water’ on basis of which load re-
ductions are estimated.
Although the APAE 2 was partly adopted in response to the Nitrates Directive (ND), the follow-
up plan, APAE 3, might offer an example of changing the 1987 goals during the APAE 3 evalua-
tion in 2008, cf. Table 3 and 7. The outcome could be seen by some as a less stringent course in
regard to losses of nitrogen from diffuse sources. Further, it seems unclear how this change fits
with the implementation of the Nitrates Directive, but also the Water Framework Directive.
42
3: Beyond Action Plans and Directives:
Perspectives for the Future
Although the Danish Action Plans and the HELCOM Baltic Sea Action Plan (BSAP) might be
considered as best practices and examples from which to learn, both plans are far from being
perfect, even though they are evidence-based.
Six issues are of interest to consider when developing a next generation of evidence-based nutri-
ent management strategies: (1) the N/P controversy, (2) so-called ‘Technical Solutions’ to abate
eutrophication, (3) target setting, (4) thresholds, (5) socio-economical aspects of eutrophication,
and (6) climate change manifested as shifting baselines.
The N/P controversy is a discussion about which nutrient input should be reduced in order to
combat eutrophication. For coastal marine waters, nitrogen has historically been considered the
limiting nutrient. However, anthropogenic phenomena affecting both sides of the N:P ratio have
combined to increase that ratio in coastal waters: Human activities have contributed to an overa-
bundance of nitrogen in coastal waters, while upstream nutrient controls focusing mainly on re-
moving P have also increased downstream N:P ratios. Schindler et al. (2008) suggested that ef-
fective eutrophication control can be achieved in both freshwater and coastal ecosystems by con-
trolling P only, based on research done in an experimental lake. Schindler et al. (2008) conclude
that fixation of atmospheric nitrogen can respond to meet ecosystem N requirements in a regime
of P enrichment, P ultimately controls eutrophication and there is no pressing need for N input
controls. Both Conley et al. (2009a) and Paerl (2009) question this finding. Evidence is presented
that both N and P must be reduced to battle eutrophication in coastal waters. Paerl (2009) points
out that nutrient dynamics in coastal and estuarine waters are quite different from those in fresh-
water systems. Coastal N2 fixation generally does not satisfy ecosystem-level N demands, caus-
ing these waters to remain N-limited and hence sensitive to N over-enrichment. HELCOM
(2009), in line with Conley et al. (2009a) and Paerl (2009), emphasize that despite regional var-
iations (the Gulf of Bothnia is P limited) control of both N and P is needed for long-term man-
agement of eutrophication in the Baltic Sea region.
The usefulness of so-called ‘technical solutions’ as an alternative to nutrient reductions in the
Baltic Sea is debatable (see Conley et al. 2009b for a summary). Proposals for technical fixes
43
include (1) artificial aeration/oxygenation, (2) large-scale manipulation of the circulation, (3)
chemical removal of phosphorus, and (4) bio-manipulation. Virtually all engineering methods
proposed to date for the Baltic Sea’s pelagic waters seem unrealistic. At best they can only speed
up recovery while nutrient reductions begin to have an effect. Conley et al. (2009b) conclude that
these large-scale attempts at remediation are unlikely to substantially improve the short-term
conditions in the Baltic Sea and several pose substantial risks for the environment. It should be
mentioned, as a precautionary note, that engineering solutions have been evaluated and rejected
(Conley et al. 2009b). However, reconstruction of stone reefs, a protected habitat type under the
EC Habitats Directive, in shallow coastal waters is likely to have a positive impact on both eu-
trophication and biodiversity (Møhlenberg et al. 2009).
A third area of concern where research is urgently required, is in regard to the setting of targets,
in particular improvement of the current understanding of and values for reference conditions
(RefCon, being the anchor of target setting) and acceptable deviation (AcDev, the acceptable
deviation from RefCon). Establishment of reference conditions in aquatic systems can be made
in a number of different ways. The methods currently used are (1) spatially based reference con-
ditions (including historical data and paleo-ecological studies), (2) modelling (empirical or dy-
namic), (3) combinations of (1) and (2), and (4) expert judgement (Andersen et al. 2004, 2010,
2011). The setting of AcDev is very critical. Experiences from the implementation of the WFD
suggest that the ‘scientific’ interpretation of the politically agreed ‘normative definitions’ (of
what an ‘acceptable’ deviation is about) has been carried out by a process not able to adequately
balance scientific advice and policy.
It is important to consider ecosystem thresholds and regime shifts as well as shifting baselines
when developing and implementing evidence-based nutrient management plans, e.g. Duarte
(2009), Duarte et al. (2009), Kemp et al. (2009), and Carstensen et al. (2011). These newly pub-
lished results are important since management of eutrophication is not only about reducing nutri-
ent losses from human activities, in particular agriculture, but also about keeping the ecological
impact of human activities at a sustainable level leading to realization of politically agreed eco-
logical objectives, e.g. a Baltic Sea ‘unaffected by eutrophication’ or ‘good ecological status in
coastal waters’.
44
Some potential implications of ecological threshold and shifting baselines in regard to eutrophi-
cation, especially in regard to reduction of loads are illustrated in Figure 12.
Panel A shows a straight forward linear recovery,
where loads have to be reduced to the loads equiva-
lent to fulfilment of the target (EutroQO = RefCon
÷ AcDev). Panel B is slightly different since it in-
cludes a threshold, meaning that the reduction re-
quired to fulfil the target is slightly larger than in a
situation without a threshold. Identification of
threshold is imperative when trying to achieve eco-
logical targets, e.g. as those in the WFD and BSAP.
Climate change will also affect the eutrophication
status of temperate coastal water in the future – and
introduce a shift in the baseline as indicated in pan-
el C – development and execution of evidence-
based nutrient management strategies ought to take
such shifts into account as soon as possible (Duarte
2009).
Combining a shifting baseline and a threshold (pan-
el D), which might be appropriate for many shallow
coastal waters, may indicate that the load reductions
required to achieve the already agreed targets might
be larger than currently acknowledged (or in ex-
treme cases impossible to achieve).
Finally, inclusion of socio-economic considerations
and cost-benefits are becoming more and more usu-
al, e.g. Wulff et al. (2007) and Anon. (2008). Swe-
den has been at the forefront of this issue, e.g.
Turner et al. (1999), Garpe (2008), and Gren &
Elofsson (2008), the latter updating the estimate of
Figure 12: Conceptual models of the con-
sequences of shifting baselines, regime shifts (thresholds) as well as the combination of shifting baseline and regime shifts for nutrient management strategies. Dashed red line indi-cated target, while the red arrow indicates required reduction of human pressures, e.g. reductions of loads. Based on Laamanen et al. (submitted).
45
the net benefits of alleviating eutrophication in the Baltic Sea. Depending on the choice of target
for nutrient reductions and choices of discount rate the overall annual net benefit ranges be-
tween 0.2 and 7.4 billion Euro per year. Such findings are very interesting, in particular since
politicians often, if not always, focus on the costs of actions. If the Swedish results are correct,
and it should be stressed that there are no reasons for questioning their methods and models, then
solving the eutrophication problem would be a good bargain. So why hasn’t it been done yet?
Probably because the Danish agriculture is better at lobbying decision-makers than environmen-
talists are. It is likely that this situation will change once national authorities and politicians be-
come informed in regard to the socio-economic consequences of continued eutrophication. The
costs of non-action are often far greater than the costs of action.
Getting more science on board when revising and developing existing nutrient management
strategies, both in terms of improved ecological understanding and better socio-economic mod-
els, will ultimately lead to more and better ecosystem-based strategies. Here, we are talking
about the fine art of balancing knowledge and economy without forgetting that policy is about
avoiding making unpopular decisions and regulations.
4: Conclusions: What Makes a Nutrient
Management Strategy Successful?
Cultural eutrophication will continue to be a significant issue for decades. However, we have
now reached a level in regard to our conceptual understanding of coastal eutrophication and its
causes where nutrient management strategies should be both informed and ecosystem-based.
With current policy drivers (WFD, MSFD, and BSAP), existing nutrient management strategies
will be updated and include more outreach and stakeholder involvement. Combining this with
the upcoming development and implementation of maritime spatial planning as well as the expe-
rience from successful nutrient management strategies, it is clear that any so-called value-based
policies aiming to hamper endeavours to reduce coastal eutrophication will be clearly exposed.
Any attempt to redefine what the problems are about should be rejected.
46
But how do we improve existing planning and make them even more evidence-based and conse-
quently ecosystem-based? Answering this question is not simple, but based on the experiences
from the two action plans discussed, an answer can be split in four ‘areas of concern’.
The first area of concern is related to political will. Generating and maintaining political will is
required for solving the problems of eutrophication in coastal waters, in particular:
‘Education’ of politicians – especially when it comes to: (1) the basic concepts and ecologi-
cal and economic consequences of eutrophication, (2) a political acceptance of the uncertain-
ties related to the estimation of load reductions, and (3) an acceptance of using the sequence
of i) Planning, ii) Action, iii) Control, and iv) Evaluation, e.g. the PACE sequence.
Goals should not to be subject to revision – the only themes to be revised by politicians are:
(1) the measures to be implemented to fulfil the goals, (2) the timing when the measures are
to be fully implemented, and (3) the overall strategy to monitor progress.
The second area of concern is related to research. If future ecosystem-based nutrient manage-
ment strategies are to be ecosystem-based, then focus ought to be put on the following themes:
Better targets – reference conditions, acceptable deviations from reference conditions, and
functional relations used for target setting should be evidence-based.
Socio-economy – a better understanding of ecosystem services and the socio-economic bene-
fits of marine waters not affected by eutrophication is required.
Habitats and species – a better understanding of the links between eutrophication, biodiversi-
ty and fisheries is required.
The third area of concern is related to long-term monitoring and assessment, which is a prerequi-
site for documenting the effect of measures and for improving the evaluation phase:
Monitoring – all ecologically relevant indicators should be monitored, spatial and temporal
coverage should be decided according to ecosystem structures and functioning, not by the
availability of funds.
Funding – should be long term, at least for 4-6 years.
The fourth area of concern is related to strategy development. Existing nutrient management
strategies should, whether they are based on the concept of either AM or EBM, be further devel-
oped and optimized. Future nutrient management strategies ought to be both adaptive and eco-
system-based. The first step should be to develop better evaluation phases of existing ecosystem-
based nutrient management strategies, a second step should be to make use of evidence-based
decision support systems linking targets, loads and costs while a third step should be to factor in
climate change.
47
It should be clear that there are many different ways leading to ecosystem-based management
taking into account differences in geographic scales, each with its own unique historical, ecolog-
ical, and social context. The future nutrient management strategies (NMS) will also vary depend-
ing on the types of legislative and managerial framework already in place. Notably, we no longer
start from scratch but need to refine the existing NMS. Perhaps most critically, the development
and implementation of any NMS cannot act in isolation. NMS’s need to be evidence-based, to
include all relevant stakeholders and to be supported by a sustained political will to alleviate eu-
trophication, in particular to reduce nutrient loads.
The DO’s and DON’T’s of evidence-based nutrient management strategies are:
DO understand that ecosystem-based management is adaptive and science-based.
DON’T assume that decisions can not be taken because of incomplete knowledge and
uncertainty.
DO evidence-based target setting and exhaustive planning, the latter involving decision-
makers, authorities and all stakeholders.
DON’T wait for perfection and all-inclusive ecosystem understanding.
DO a full execution of the plan.
DON’T rely on voluntary agreements or guidelines.
DO monitoring with ecologically relevant resolution in time and space.
DON’T underestimate resources needed for sampling, quality assurance, analysing data and
reporting.
DO regular evaluations in regard to the progress of the nutrient management strategy.
DON’T disregard the advantages of a dual monitoring strategy focusing on both nutrient
inputs as well as ecological responses to lowered nutrient inputs.
It is critical to be patient as emphasised by Fulweiler et al. (2010) since time is needed before the
effects of changes in human behaviour can be seen in inputs and eventually in the ecological
quality of the marine environment. It is also critical to prepare for those moments where deci-
sions and actions can be taken by building up the best possible scientific basis for decision-
making as well as decision support systems. An important lesson learned from the Danish Action
Plans on the Aquatic Environment and the HELCOM Baltic Sea Action Plan is that things come
about in windows of opportunity.
48
5: References
Ærtebjerg Nielsen, G., T.S. Jacobsen, E. Gargas & E. Buch, 1981: The Belt Project. Evaluation
of the Physical, Chemical and Biological Measurements. The National Agency of Envi-
ronmental Protection, Denmark. 122 pp.
Ærtebjerg, G., J.H. Andersen & O.S. Hansen, 2003: Nutrients and Eutrophication in Danish Ma-
rine Waters. A Challenge to Science and Management. National Environmental Research
Institute. 126 p. Available at:
http://www2.dmu.dk/1_Viden/2_Publikationer/3_ovrige/rapporter/Nedmw2003_alle.pdf
Andersen, J.H., 2010: Eutrophication. Baltic Sea Environmental Proceedings 122:16-17. In:
HELCOM, 2010: Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assess-
ment 2003-2007. Baltic Sea Environment Proceedings 122. 63 pp
Andersen, J.H., & D.J. Conley, 2009: Coastal marine eutrophication: towards better understand-
ing and nutrient management strategies. Hydrobiologia 621(1):1-4.
Andersen, J.H., D.J. Conley & S. Hedal, 2004: Palaeo-ecology, reference conditions and classifi-
cation of ecological status: the EU Water Framework Directive in practice. Marine Pollu-
tion Bulletin 49:283-290.
Andersen, J.H., L. Schlüter & G. Ærtebjerg, 2006: Coastal eutrophication: Recent developments
in definitions and implication for monitoring strategies. Journal of Plankton Research
28(7):621-628.
Andersen, J.H., P. Axe, H. Backer, J. Carstensen, U. Claussen, V. Fleming-Lehtinen, M. Jä-
rvinen, H. Kaartokallio, S. Knuuttila, S. Korpinen, A. Kubiliute, M. Laamanen, E. Lysiak-
Pastuszak, G. Martin, F. Møhlenberg, C. Murray, G. Nausch, A. Norkko, & A. Villnäs,
2010a: Getting the measure of eutrophication in the Baltic Sea: towards improved assess-
ment principles and methods. Accepted by Biogeochemistry.
Andersen, J.H., C. Murray, H. Kaartokallio, P. Axe & J. Molvær, 2010b: Confidence rating of
eutrophication status classification. Marine Pollution Bulletin 60:919-924.
Andersen, J.H. & J. Carstensen, 2011: Gisp. Vandmiljøplaner har været en succes.
Analyseartikel i Politiken Debat den 15. oktober 2011, p. 4. (in Danish)
Anon., 1991a: Council Directive of 21 May 1991 concerning urban waste water treatment
(91/271/EEC). Official Journal of the European Communities L 135.
Anon., 1991b: Council Directive 91/676/EEC of 12 December 1991 concerning the protection of
waters against pollution caused by nitrates from agricultural sources. Official Journal of
the European Communities L 375.
Anon., 2000: Directive 200/60/EC of the European Parliament and of the Council of 23 October
2000 establishing a framework for Community action in the field of water policy. Official
Journal of the European Communities L 327/1.
Anon., 2004: Agreement on the Action Plan for the Aquatic Environment III 2005-2015 between
the Danish Government, the Danish People's Party and the Christian Democrats. Viewed
22 July 2009 at:
http://www.vmp3.dk/Files/Filer/English%20version/engelsk_oversaettelse.pdf
49
Anon., 2008a: Gulf Hypoxia Action Plan for reducing, mitigating, and controlling hypoxia in the
Northern Gulf of Mexico and improving water quality in the Mississippi River Basin. 62
pp.
Anon., 2008b: Directive 2008/56/EC of the European Parliament and of the Council of 17 June
2008 establishing a framework for community action in the field of marine environmental
policy (Marine Strategy Framework Directive). Official Journal of the European Commu-
nities L 164/19.
Anon., 2009b: Guidance document on eutrophication assessment in the context of European wa-
ter policies. Guidance Document No. 23. Technical Report - 2009-030. 136 pp.
ATV, 1990: Vandmiljøplanens tilblivelse og iværksættelse. Akademiet for de tekniske Viden-
skaber. ISBN 87-88699-74-9. 152 pp. (in Danish)
Bachmann, R.W., J.E. Cloern, R.E. Hecky & D.W. Schindler (eds.), 2006: Eutrophication in
Freshwater and Marine Systems. Special Issue. Limnology & Oceanography 51:351-800.
Backer, H., 2008: Indicators and scientific knowledge in regional Baltic Sea environmental poli-
cy. ICES Journal of Marine Science 65:1398-1401. DOI:10.1093/icesjms/fsn157
Backer, H. & J.M. Leppänen, 2008: The HELCOM system of a vision, strategic goals and eco-
logical objectives: implementing an ecosystem approach to the management of human ac-
tivities in the Baltic Sea. Aquatic Conservation 18(3):321-334. DOI: 10.1002/aqc.8
Backer, H., J.M. Leppänen, A.C. Brusendorff, K. Forsius, M. Stankiewicz, J. Mehtonen, M. Py-
hälä, M. Laamanen, H. Paulomäki, N. Vlasov & T. Haaranen, 2010: HELCOM Baltic Sea
Action Plan – A regional programme of measures for the environment based on the Eco-
system Approach. Marine Pollution Bulletin 60(5):642-649.
Bosch, D.F., R.B. Brinsfield & R.E. Magnien, 2001: Chesapeake Bay Eutrophication: Scientific
Understanding, Ecosystem Restoration, and Challenges for Agriculture. J. Environ. Qual.
30:303-320.
Bosch, D.F., 2006: Scientific requirements for ecosystem-based management in the restoration
of Chesapeake Bay and Coastal Louisiana. Ecological Engineering 26:6-26.
Carstensen, J., D.J. Conley & P. Henriksen, 2004: Frequency, composition, and causes of sum-
mer phytoplankton blooms in a shallow coastal ecosystem, the Kattegat. Limnology and
Oceanography 49(1):190-201.
Carstensen, J., D.J. Conley, J.H. Andersen & G. Ærtebjerg, 2006: Coastal eutrophication and
trend reversal: A Danish case study. Limnology & Oceanography 51(1, part 2):398-408.
Carstensen, J., M. Sánchez-Camacho, C.M. Duarte, D. Krause-Jensen & N. Marba, 2011: Con-
necting the Dots: Responses of Coastal Ecosystem to Changing Nutrient Concentrations.
Environmental Science and Technology 45:9122-9132.
Christensen, N.L., A. Bartuska, J.H. Brown, S. Carpenter, C. D'Antonio, R. Francis, J.F. Frank-
lin, J.A. MacMahon, R.F. Noss, D.J. Parsons, C.H. Peterson, M.G. Turner & R.G. Mood-
mansee, 1996: The report of the Ecological Society of America Committee on the scien-
tific basis for ecosystem management. Ecological Applications 6:665-691.
Christensen, P.B., F. Møhlenberg, L.C. Lund-Hansen, J. Borum, C. Christiansen, S.E. Larsen,
M.E. Hansen, J. Andersen & J. Kirkegaard, 1998: The Danish Marine Environment: Has
Action Improved its State? - Havforskning fra Miljøstyrelsen, nr. 62. 115 p. Available at:
http://www2.mst.dk/Udgiv/publications/1998/87-7810-967-1/pdf/87-7810-967-1.pdf
50
Cloern, J., 2001: Our evolving conceptual model of the coastal eutrophication problem. Marine
Ecology Progress Series 210:223-253.
Conley, D.J., H. Kaas, F. Møhlenberg, B. Rasmussen & J. Windolf, 2000: Characteristics of
Danish estuaries. Estuaries 23(6):820-837.
Conley, D.J., S. Markager, J.H. Andersen, T. Ellermann & L.M. Svendsen, 2002: Coastal Eu-
trophication and the Danish National Aquatic Monitoring and Assessment Program. Estu-
aries 25:848-861.
Conley, D.J., J. Carstensen, G. Ærtebjerg, P.B. Christensen, T. Dalsgaard, J.L.S. Hansen & A.B.
Josefson, 2007: Long-term changes and impacts of hypoxia in Danish coastal waters. Eco-
logical Applications 17:165-184.
Conley, D.J., H.W. Paerl, R.W. Howarth, D.F. Boesch, S.P. Seitzinger, K.E. Havens, C. Lancelot
& G.E. Likens, 2009a: Controlling Eutrophication: Nitrogen and Phosphorus. Science
323:1014-1015.
Conley, D. J., E. Bonsdorff, J. Carstensen, G. Destouni, B.G. Gustafsson, L.-A. Hansson, N.N.
Rabalais, M. Voss & L. Zillén, 2009b: Tackling hypoxia in the Baltic Sea: Is engineering a
solution? Environmental Science and Technology 43:3407-3411.
Diaz, R.J. & R. Rosenberg, 2008: Spreading dead zones and consequences for marine ecosys-
tems. Science 321:926-929.
Duarte, C.M., 2009: Coastal eutrophication research: a new awareness. Hydrobiologia 629:263-
269. DOI: 10.1007/s10750-009-9795-8.
Duarte, C.M., D.J. Conley, J. Carstensen & M. Sánchez-Camacho, 2009: Return to Neverland:
Shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32:29-36.
DOI: 10.1007/s12237-008-9111-2.
Elliot, M., & V.N. de Jonge, 2002: The management of nutrients and potential eutrophication in
estuaries and other restricted water bodies. Hydrobiologia, 475.476:513-524.
Elliot, M., T.F. Fernandes V.N. de Jonge, 1999: The impact of European Directives on estuarine
and coastal science and management. Aquatic Ecology 33:311-321.
Elmgren, R., 2001: Understanding human impact on the Baltic ecosystem: changing view in re-
cent decades. Ambio 30:222-231.
Ernst, H.R., 2010: Fight for the Bay. Why a Dark Green Environmental Awakening Is Needed to
Save the Chesapeake Bay. Rowman & Littlefield Publishers, Inc. 145 pp.
Fallesen, G., F. Andersen & B. Larsen, 2000: Life, death and revival of the hypertrophic Ma-
riager Fjord, Denmark. Journal of Marine Systems 25(3-4):313-321. DOI: 10.1016/S0924-
7963(00)00024-5.
Foden, J., S.I. Rogers & A.P. Jones, 2008: A critical review of approaches to aquatic environ-
mental assessment. Marine Pollution Bulletin 56:1825-1833.
DOI:10.1016/j.marpolbul.2008.08.017.
Folketinget 1984-85: Folketingsbeslutning om nedbringelse af forureningen med næringsssalte
og organisk stof (NPo handlingsplanen). Blad nr. 937. 2 pp (In Danish)
Folketinget, 1986-1987: Beretning over vandmiljøplanen. Beretning afgivet af miljø- og plan-
lægningsudvalget den 30. april 1987. Blad nr. 817. 12 pp (in Danish)
51
Fulweiler, R.W., A-S. Heiskanen & N.N. Rabalais, 2010: The Eutrophication Commandments.
EUTRO 2010 Summary Poster presented at the OSPAR Ministerial Meeting in Bergen,
Norway. Available via http://www.ices.dk/iceswork/symposia/eutro2010posterFinal.pdf.
Garpe, K., 2008: Ecosystem services provided by the Baltic Sea and Skagerrak. Report 5873
from the Swedish Environmental Protection Agency. 191 pp.
Grant, R., & G. Blicher-Mathiesen, 2004: Danish policy measures to reduce diffuse nitrogen
emissions from agriculture to the aquatic environment. Water Science and Technology
49(3):91-100.
Grant, R., & J. Waagepetersen, 2003: Vandmiljøplan II - slutevaluering. Danmarks
Miljøundersøgelser. 32 pp.
Grant, R., K. Nielsen & J. Waagepetersen, 2006: Reducing Nitrogen Loading of Inland and Ma-
rine Waters - Evaluation of Danish Policy Measures to Reduce Nitrogen Loss from Farm-
land. Ambio 35(3):117-123.
Gren, I.M., & K. Elofsson, 2008: Costs and benefits from nutrient reductions to the Baltic Sea.
Report 5877 from the Swedish Environmental Protection Agency. 67 pp.
Halpern, B.S., S. Walbridge, K.A. Selkoe, C.V. Kappel, F. Micheli, C. D’Agrosa, J.F. Brunoa, et
al., 2008: A global map of human impact on marine ecosystems. Science 319:948-952.
HELCOM, 1988: Declaration of the Protection of the Marine Environment of the Baltic Sea Ar-
ea.6 pp. Available at:
http://www.helcom.fi/stc/files/MinisterialDeclarations/MinDecl1988.pdf
HELCOM, 2005: HELCOM Monitoring and Assessment Strategy. Annex 3 of the Minutes of
the 26th Meeting of the Baltic Marine Environment Protection Commission (HELCOM
26/2005). Available at: http://www.helcom.fi/groups/monas/en_GB/monitoring_strategy/
HELCOM, 2006: Development of tools for assessment of eutrophication in the Baltic Sea. Baltic
Sea Environmental Proceedings No. 104. 64 pp.
HELCOM, 2007: HELCOM Baltic Sea Action Plan. Helsinki Commission. 103 pp. Available at:
http://www.helcom.fi/BSAP/en_GB/intro/
HELCOM, 2008: Manual for the Marine Monitoring in the COMBINE Programme of HEL-
COM. Available at: http://www.helcom.fi/groups/monas/CombineManual/en_GB/Contents
HELCOM, 2009: Eutrophication in the Baltic Sea. An integrated thematic assessment of eu-
trophication in the Baltic Sea region. Eds.: J.H. Andersen & M. Laamanen. Baltic Sea En-
vironmental Proceedings No. 115B. Helsinki Commission. 148 pp. Available at:
http://meeting.helcom.fi/c/document_library/get_file?p_l_id=79889&folderId=377779&na
me=DLFE-36818.pdf
HELCOM, 2010: Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment
2003-2007. Baltic Sea Environment Proceedings 122. 63 pp
HELCOM & OSPAR, 2003: Declaration of the first joint ministerial meeting of the Helsinki and
OSPAR Commissions. Available at:
http://www.helcom.fi/ministerial_declarations/en_GB/ospardeclaration/
Hovedstadsrådet, 1983: Teknisk baggrundsnotat, 1 Roskilde Fjord og opland, recipientgrundlag,
kvalitetskrav og konsekvensberegninger 1982. In: Forslag til Recipientkvalitetsplan og
spildevandsplan for Roskilde Fjord og opland. Planlægningsrapport nr. 36, 60 pp. (in Da-
nish)
52
Indenrigsministeriet, 1988: Vandmiljøplanens overvågningsprogram, akt nr. 45 af 13. November
1987. 5 pp.
Iversen, T.M., R. Grant & K. Nielsen, 1998: Nitrogen enrichment of European inland and marine
waters with special attention to Danish policy measures. Environmental Pollution
102(1/1):771-780.
Jensen, P.N, S. Boutrup, L. van der Bijl, L.M. Svendsen, R. Grant, P. Wiberg-Larsen, T.B.
Jørgensen, T. Ellermann, M. Hjorth, A.B. Josefson, M. Bruus, B. Søgaard, L. Thorling &
K.D. Laursen, 2010: Vandmiljø og Natur 2008. NOVANA. Tilstand og udvikling – faglig
sammenfatning. – Faglig rapport fra DMU nr. 767. 106 pp (In Danish)
Jørgensen, B.B., & K. Richardson (eds.), 1996: Eutrophication in Coastal Marine Ecosystems.
Coastal and Estuarine Studies, 52. American Geophysical Union. Washington, DC. 273 p.
Josefson, A.B. & J.L.S. Hansen, 2004: Species richness of benthic macrofauna in Danish estuar-
ies and coastal areas. Global Ecology and Biogeography 13:273-288.
Kemp, W.M., W.R. Boynton, J.E. Adolf, D.F. Boesch, W.C. Boicourt, G. Brush, J.C. Cornwell,
T.R. Fisher, P.M. Glibert, J.D. Hagy, L.W. Harding, E.D. Houde, D.G. Kimmel, W.D. Mil-
ler, R.I.E. Newell, M.R. Roman, E.M. Smith, J.C. Stevenson, 2005: Eutrophication of
Chesapeake Bay: historical trends and ecological interactions. Marine Ecology Progress
Series 303:1-29.
Kemp, W.M., J.M. Testa, D.J. Conley, D. Gilbert & J. D. Hagy, 2009: Temporal responses of
coastal hypoxia responses to loading and physical control. Biogeosciences 6:2985-3008.
Kononen, K., & E. Bonsdorf (eds.), 2001: Man and the Baltic Sea. Special Issue. Ambio 30:171-
326.
Korpinen, S., L. Meski, J.H. Andersen & M. Laamanen, 2012: Human pressures and their po-
tential impact on the Baltic Sea ecosystem. Ecological Indicators 15:105-114.
Kronvang, B., G. Ærtebjerg, R. Grant, P. Kristensen, M. Hovmand & J. Kirkegaard, 1993: Na-
tionwide Monitoring of Nutrients and Their Ecological Effects: State of the Danish Aquat-
ic Environment. Ambio 22(4):176-187.
Kronvang, B., H.E. Andersen, C. Børgesen, T. Dalgaard, S.E. Larsen, J. Bøgestrand & G. Bli-
cher-Mathiasen, 2008: Effects of policy measures implemented in Denmark on nitrogen
pollution of the aquatic environment. Environmental Science & Policy 11(2):144-152.
Laamanen, M., S. Korpinen, U.-L. Zweifel & J.H. Andersen, submitted: Ecosystem health. Text-
book chapter in “Biological Oceanography of the Baltic Sea” (Eds: P. Snoeijs, H. Schubert
& T. Radziejewska).
Läine, A., H. Pitkänen, B. Arheimer, H. Behrendt, W. Jarosinski, S. Lucane, K. Pachel, A.
Räike, A. Shekhovtsov, L.M. Svendsen & S. Valatka, 2002: Evaluation of the implementa-
tion of the 1988 Ministerial Declaration regarding nutrient load reductions in the Baltic Sea
catchment area. The Finnish Environment 524. 195 pp.
McLeod, K.L., J. Lubchenco, S.R. Palumbi & A.A. Rosenberg, 2005: Scientific Consensus
Statement on Marine Ecosystem-Based Management. Signed by 217 academic scientists
and policy experts with relevant expertise and published by the Communication Partner-
ship for Science and the Sea at http://compassonline.org/?q=EBM.
McLeod, H.M., & K.L. Leslie, 2007: Confronting the challenges of implementing marine eco-
system-based management. Front Ecol Environ 5(10):540-548. DOI:10.1890/060093.
53
Miljøministeriet, 1987: Handlingsplan mod forurening af det danske vandmiljø med næringssal-
te. 13 pp. (in Danish)
Miljøstyrelsen, 1984a: Iltsvind og fiskedød i 1981. Omfang og årsager. Miljøstyrelsen. 247 pp
(In Danish)
Miljøstyrelsen, 1984b: NPo redegørelsen. 218 pp. (In Danish)
Miljøstyrelsen, 1989: Vandmiljøplanens overvågningsprogram. Miljøprojekt nr. 115.
Miljøstyrelsen. 64 pp (In Danish)
Miljøstyrelsen, 1990: Vandmiljø-90. Redegørelse fra Miljøstyrelsen nr. 1, 1990. 204 pp (In Da-
nish)
Miljøstyrelsen, 1993: Vandmiljøplanens overvågningsprogram 1993-1997. Redegørelse fra Mil-
jøstyrelsen nr. 2, 1993. 172 pp. (In Danish)
Miljøstyrelsen, 2000: NOVA-2003. Programbeskrivelse for det nationale program for overvåg-
ning af vandmiljøet i Danmark, 1998-2003. Redegørelse fra Miljøstyrelsen nr. 1, 2000. 198
pp. (In Danish)
Møhlenberg, F., J.H. Andersen (eds.), C. Murray, P.B. Christensen, T. Dalsgaard, H. Fossing &
D. Krause-Jensen (2008): Stenrev i Limfjorden: Fra naturgenopretning til supplerende vir-
kemiddel. DHI Teknisk Rapport til By- og Landskabsstyrelsen. 41 p + bilag. (In Danish)
Nielsen, S.L., K. Sand-Jensen, J. Borum & O. Geertz-Hansen, 2002a: Depth colonisation of eel-
grass (Zostera marina) and macroalgae as determined by water transparency in Danish
coastal waters. Estuaries 25:1025-1032.
Nielsen, S.L., K. Sand-Jensen, J. Borum & O. Geertz-Hansen, 2002b: Phytoplankton, nutrients
and transparency in Danish coastal waters. Estuaries 25:930-937.
Nixon, S.W., 1995: Coastal marine eutrophication: a definition, social causes, and future con-
cerns. Ophelia 41:199-219.
Nordemann Jensen et al. (2010): Vandmiljø og nature 2008. NOVANA. Tilstand og udvikling –
faglig sammenfatning. Faglig rapport fra DMU nr. 767. 106 pp (In Danish)
OSPAR, 2008: Second integrated report on the eutrophication status of the OSPAR maritime
area. Eutrophication Series. OSPAR Commission. 107 pp
Paerl, H.W., 2009. Controlling eutrophication along the freshwater-marine continuum: Dual nu-
trient (N and P) reductions are essential. Estuaries and Coasts 32:593-601. DOI:
10.1007/s12237-009-9158-8.
PH-Consult ApS, 1993: Spildevandsforskning 1987-1992. Spildevandsforskning fra Miljøstyrel-
sen nr. 53. 337 pp. (In Danish)
Rabalais, N.N. & S.W. Nixon (eds.), 2002: Nutrient Over-enrichment in Coastal Waters: Global
Patterns of Cause and Effect. Dedicated Issue. Estuaries 25:639-900.
Rabalais, N.N., R.E. Turner & D. Scavia (2002): Beyond science into policy: Gulf of Mexico
hypoxia and the Mississippi River. BioScience 52(2):129-142.
Savchuk, O.P., & F. Wulff, 2007: Modelling the Baltic Sea eutrophication in a decision support
system. Ambio 36:141-148.
Schindler, D.W., R.E. Hecky, D.L. Findlay, M.P. Stainton, B.R. Parker, M. Paterson, K.G.
Beaty, M. Lyng, & S.E.M. Kasian, 2008: Eutrophication of lakes cannot be controlled by
54
reducing nitrogen input: Results of a 37 year whole ecosystem experiment. Proceedings of
the National Academy of Science 105:11254-11258. DOI: 10.1073/pnas.0805108105.
Suman, D., S. Guerzoni & E. Molinaroli, 2005: Integrated coastal management in the Venicela-
goon and its watershed. Hydrobiologia 550(1):251-269.
Svendsen L.M., L. van der Bijl, S. Boutrup & B. Norup (eds.), 2005: NOVANA: Nationwide
Monitoring and Assessment Programme for the Aquatic and Terrestrial Environments.
Programme Description - Part 2. National Environmental Research Institute, Denmark. -
NERI Technical Report No. 537. 137 pp.
Turner, R.K., S. Georgiou, I.M. Gren, F. Wulff, S. Barrett, T. Söderqvist, I.J. Bateman, C. Folke,
S. Langaas, T. Zylick, K.G. Mäler & A. Markowska, 1999: Managing nutrient fluxes and
pollution in the Baltic: an interdisciplinary simulation study. Ecological Economics
30:333-352.
Waagepetersen, J., R. Grant, C.D. Børgesen & T.M. Iversen, 2008: Midtvejsevaluering af
Vandmiljøplan III. Det Jordbrugsvidenskabelige Fakultet, Århus Universitet og Danmarks
Miljøundersøgelser, Århus Universitet. 36 pp. (In Danish)
Wulff, F., O.P. Savchuk, A.V. Sokolov, C. Humborg & M. Mörth, 2007: Assessing the past and
the possible future of the Baltic. Ambio 36:243-249.
Sources of unpublished information
Kristoffer Colding, pers. comm., the Danish Nature Agency, Ministry for the Environment,
Haraldsgade 53, 2100 Copenhagen Ø, Denmark
Jens Brøgger Jensen, pers. comm., the Danish Nature Agency, Ministry for the Environment,
Haraldsgade 53, 2100 Copenhagen Ø, Denmark
Maria Laamanen, pers. comm., HELCOM Secretariat, Katajanokanlaituri 6B, FIN-00160
Helsinki, Finland
55
Annex 1: Abstract
Nutrient management strategies have to deal with all human activities resulting in discharges and
losses of nutrients, e.g. land-use, fertiliser use, industrial production, households and energy con-
sumption.
Two Nutrient Management Strategies, one national based on Adaptive Management (AM) and
one trans-national based on an Ecosystem-Based Approach to management of human activities,
in practise being equivalent to Ecosystem-Based Management (EBM), are analysed.
The aim of this thesis is to analyze the critical factors likely to be required for a successful man-
agement strategy. Obviously, two key factors are nutrient reductions, generally of both nitrogen
and phosphorus, and monitoring of environmental status and trends.
It is likely that the prescribed quantity of measures ( e.g. nutrient reductions) is not a critical fac-
tor as long as periodic evaluation of progress and publication of assessments play a strong role in
the strategy, especially in combination with a sustained political will to follow up on the evalua-
tions and pursue the visions and objectives of the strategy in question.
A combination of AM and EBM should be advanced to improve evidence-based nutrient man-
agement strategies.
56
Annex 2: Nutrients Discharges and Losses in Denmark
1989-2008
Table A2.1
Discharges of nitrogen and phosphorus (in tonnes) from urban waste water treatment plants
1989-2008.
Year Nitrogen (tot N) Phosphorus (tot P)
1989 18,000 4,470
1990 16,900 3,710
1991 15,100 2,800
1992 13,100 2,260
1993 10,800 1,760
1994 10,200 1,570
1995 8,900 1,230
1996 6,390 900
1997 4,850 670
1998 5,170 600
1999 5,130 580
2000 4,650 540
2001 4,220 470
2002 4,530 510
2003 3,610 400
2004 4,030 430
2005 3,810 410
2006 3,610 390
2007 4,360 470
2008 3,550 460
57
Table A2.2
Discharges of nitrogen and phosphorus (in tonnes) from industries with separate discharge 1989-
2008.
Year Nitrogen (tot N) Phosphorus (tot P)
1989 6,500 1,410
1990 4,080 650
1991 3,770 520
1992 4,180 410
1993 2,540 240
1994 2,680 310
1995 2,440 200
1996 1,790 130
1997 1,760 130
1998 1,350 120
1999 970 70
2000 900 60
2001 820 50
2002 760 50
2003 510 30
2004 70 30
2005 440 20
20061 - -
2007 320 20
2008 400 20
1: Discharges from 2006 have unfortunately not been estimated as a significant part of the pri-
mary data from individual industries with separate discharges has not been reported. The reason
is the Structural Reform implemented by 1 January 2007, especially the cessation of the coun-
tries in Denmark.
58
Table A2.3
Nitrogen surplus (in tonnes) in Danish agriculture 1980-2008.
Year Nitrogen (tot N)
1980 434,000
1981 394,000
1982 381,000
1983 453,000
1984 400,000
1985 420,000
1986 408,000
1987 423,000
1988 376,000
1989 374,000
1990 379,000
1991 396,000
1992 431,000
1993 360,000
1994 356,000
1995 320,000
1996 303,000
1997 290,000
1998 292,000
1999 278,000
2000 267,000
2001 260,000
2002 247,000
2003 224,000
2004 229,000
2005 215,000
2006 188,000
2007 206,000
2008 209,000
59
Table A2.4
Losses of nitrogen (tot N) from the root zone of sandy soils and clay soils 1990/91 to 2007/08.
Year Sandy soils Clay soils Weighted mean Weighted reduction
(kg N/ha) %
1990/91 154 76 107 0
1991/92 144 72 101 5,6
1992/93 139 68 96 10,3
1993/94 129 64 90 15,9
1994/95 118 66 87 18,7
1995/96 109 60 80 25,2
1996/97 102 58 76 29,0
1997/98 101 60 76 29,0
1998/99 83 53 64 40,2
1999/2000 84 51 64 40,2
2000/01 84 51 64 40,2
2001/02 82 49 62 42,1
2002/03 80 47 60 43,9
2003/04 81 46 60 43,9
2004/05 85 47 62 42,1
2005/06 83 43 59 44,9
2006/07 85 47 62 42,1
2007/08 88 46 63 41,1
The weighted mean is based on Børgesen, C.D. & R. Grant, (2003): Baggrundsnotat til Vandmil-
jøplan II - slutevaluering. Vandmiljøplan II - modelberegning af kvælstofudvaskning på lands-
plan, 1984-2002. Danmarks JordbrugsForskning og Danmarks Miljøundersøgelser. 22 pp, which
is available via:
http://www.agrsci.dk/var/agrsci/storage/original/application/phpE3.tmp.pdf.
60
Annex 3: Curriculum Vitae for Jesper H. Andersen
Name Jesper Harbo Andersen Date of birth May 4, 1962 Nationality Danish Education
Courses
M.Sc. (Aquatic Ecology), University of Copenhagen (1990) 2006: Introduction to Microsoft Project (MaCom A/S) 2004: Introduction to Management (DIEU) 2002: Coaching (CONFEX) 2002: How to write a Technology Implementation Plan (Hyperion Ltd., Ireland) 2000: Teambuilding (DME) 1999: Communication and collaboration (DIEU) 1997: English Language Performance Training Programme (Danida) 1997: Course on Design of Water Quality Monitoring Networks (VKI and ColStat) 1996: Logical Framework Approach (COWI) 1995: Meetings and Negotiations – A Seminar in English (FLEX-SPROG) 1991: Crayfish Management and Culture (University of Kuopio, Finland)
Key qualifications
Jesper H. Andersen’s primary fields of interest are aquatic ecology, development of assessment tools (multi-metric and indicator-based), design and planning of monitoring networks (and coupled courses of action regard-ing data flow, QA, data storage and reporting), ecosystem-based management as well as feeding scientific information into adaptive and evidence-based management processes. He has a comprehensive knowledge of aquatic ecology with special emphasis on impacts of human activities, environmental regulatory processes and their administrative and economic contexts.
Jesper H. Andersen has broad experiences in relation to management of teams and large projects and is Project Director at Institute of Bioscience, Aarhus University. He coordinates the institutes work in relation to the EU Marine Strategy Framework Directive. Recently, he has been national expert in the MSFD Eutrophication Task Group, the Danish Marine Strategy Expert Network as well as chair of the HELCOM HOLAS Task Force. He has earlier on worked as a national expert in the WFD CIS COAST group and the WFD CIS Eutrophication Activity. Another key qualification is organisation of international conferences, e.g. EUTRO 1993, the Symposium on the North Sea QSR in 1994, EUTRO 2006, the ICES Indicator Symposium in 2007, EUTRO 2010 and the upcoming Marine Strategy 2012 Conference during the upcoming Danish EU Presidency.
He has been heading of the Danish National Marine Monitoring Centre (M-FDC) in 2001-2005, which co-ordinates the national marine monitoring and assessment program (then: 18 partners and an annual budget of ~7.0 mio. €). He has been Danish HOD in the HELCOM Environmental Committee (EC, now MONAS) and the OSPAR ASMO and worked in many other international groups. Further, he has participated in EU funded RTD projects (CHARM, MOLTEN, DANLIM, EUSeaMap) and EEA’s ETC Water 2007-2010.
Project manager of large projects, such as: Danish EPA projects on the Water Framework Directive 1998-2004 (phase I, IIa, IIb, III and IV), NMR RETRO project 2002-2004, HELCOM EUTRO-PRO 2005-2008, BALANCE 2005-2007 (nominated for the UN ENERGY GLOBE Award 2009 and selected by the jury as national winner of ENERGY GLOBE Award 2009 and also by WWF rated as one out of five 2008 high-lights in the Baltic Sea area), HELCOM HOLAS 2009-2010 and HARMONY 2010-2012.
Jesper H. Andersen’s list of publications includes scientific papers, technical reports, policy papers, books, news paper articles and includes more than 135 references.
Memberships Chairman, Committee on Public Sector Consultancy and Applied Research, AU BioScience (2012 - ) Expert member of the National Nature Protection and Environmental Board of Appeal 2011-2014
Member, Advisory Board of the FORMAS project “Managing Multiple Stressors in the Baltic Sea” Coastal and Estuarine Research Federation (CERF – formerly ERF) Baltic Marine Biologists (BMB) Danish Society for Environmental Engineering (IDAmiljø)
Teaching Lecturing in: “Systems Ecology”, Roskilde University Centre (RUC), focusing on nutrient enrichment of marine waters and
adaptive ecosystem-based management in an international and national perspective. Duration: 2003–ongoing. “The Baltic Sea: Yesterday, Today and Tomorrow”, Ph.D. course at Lund University. Duration: 2009–ongoing. “Environment and Resources”, Danish Technical University. Duration: 2006–2007. “Freshwater Ecology”, Copenhagen University. Duration: 2005–2006.
61
Employment record Year Firm Position and responsibilities 2011 –
Department of Bioscience, Aarhus University (formerly Department of Marine Ecology, National Environmental Research Institute at Aarhus University)
Jesper H. Andersen is employed as projektchef (Project Director) at Department of Bioscience at Aarhus University. He is currently chairman of the institute’s Committee on Public Sector Consultancy and Applied Research. Projects under negotiation and/or development:
DEVOTES, a European MSFD FP7 research project focusing on GES (indicators, modelling, targets), sea-based and land-based pressures, a suite of case studies (e.g. the Kattegat) and guidance in regard to ecosys-tem-based management strategies. Planned budget: 9 mio. €.
SYMBIOSE, a national research and development in support of the Marine Strategy Framework Directive. Planned budget: 4.,5 mio. dkr.
MONET, phase 2 (Development of innovative methods for monitoring and design of monitoring networks for characterisation of Baltic Sea ecosys-tems), an application for BONUS. Planned budget: 4 mio. €.
HARMONY, phase 3, specific spin-off activities from the HARMONY project Planned budget: 45.000 €.
Ongoing projects and supervision:
Project partner in HELCOM TARGREV, phase 2 (2012), focusing on
revision and publication of the HELCOM TARGREV project. Budget: 15.000 euro.
Project partner in Baltic NEST Institute (2012), contribution to specific task
related to ecosystem-based management of eutrophication in the Baltic Sea. Budget: to be decided.
Project manager of HARMONY, phase 2 (2011-2012), the continuation and
finalisation of the HARMONY project initiated in 2010. Focus in on the de-velopment of tools for 1) indicator-based assessment of ‘good enviro-nmental status’ and cumulative anthropogenic pressures in the North Sea. Budget, phase 2: 145.000 €.
MONET, phase 1 (Development of innovative methods for monitoring and
design of monitoring networks for characterisation of Baltic Sea ecosys-tems), an application for BONUS. Budget: 200.000 dkr.
Conference Secretary for Marine Strategy 2012, a three day conference
(14-16 May 2012) during the Danish EU Presidency focusing on research and ecosystem-based management strategies in support of the EU Marine Strategy Framework Directive. Budget: > 3 mio. dkk.
Project partner in Review of Femerbelt EIA (2011-2012), a a review of the
draft Environmental Impact Assessment for the fixed link in Femerbelt. Budget: Confidential.
Project partner in WATERS (2011-2015), a Swedish RDI project aiming to
develop indicators and assessment tools in regard to the EU Water Frame-work Directive. Budget, NERI: 2,3 mio dkk.
62
Year Firm Position and responsibilities
Completed projects and supervision from January 2011:
Faglige baggrundsnotater til havstrategidirektivets basisanalyse (“Havet omkring Danmark”, phase 2), drafting of specific technical contribu-tions to the Danish initial assessments pursuant to the EU Marine Strategy Framework Directive. Budget: 1 mio. dkk.
Sub-consultant in HELCOM TARGREV, phase 1 (2010-2011), a specific
NERI contribution to DHI in order to fulfil DHI’s commitments under the HELCOM TARGREV project. Budget for sub-contract: 29.376 euro.
Sub-consultant in BWO (2010), a specific NERI contribution to DHI in order
to fulfil DHI‘s commitments under the BWO Interreg project. Budget for sub-contract, phase 1: 5.000 euro. Budget for phase 2: 45.000 euro.
Sub-consultant in MARCOS (A NMR funded project focusing on “Marine
European Directives: Concepts, Overlap and Synergies”), a specific NERI contribution to DHI in order to fulfil DHI’s commitments under the MARCOS project. Budget for sub-contract: 140.000 dkr.
Project manager of “MSFD synopsis” (phase 1), drafting of a synopsis for
the Danish initial assessments pursuant to the EU Marine Strategy Frame-work Directive. Budget: 75.000 dkk.
2005 – 2010 DHI Water Environment Health Head of EU Water Policy Team, Department of Ecology & Environment: business
area manager for DHI’s activities in relation to the EU Water Framework Directive and the EU Marine Strategy Framework Directive as well as key account manag-er. Number of staff implicated: 14 persons.
Projects and supervision in the period 2005-2010:
Project partner in Service contract for support to the implementation of the Marine Strategy Framework Directive, a tender published by DG ENV. Total budget: 450.000 € per year (2010-2013).
HARMONY, phase 1 (2010); a Danish, Norwegian and Swedish RDI pro-
ject aiming to develop tools for initial assessments cf. the EU Marine Strate-gy Framework Directive. Budget, phase 1: 100.000 euro
HELCOM TARGREV (2010-2011), a RDI project focusing on updating the
eutrophication segment and the country-wise load allocations of the Baltic Sea Action Plan. Budget: 63.000 €.
Project partner in the EEA ETC/ICM for 2011-2014 with focus on marine
and maritime tasks.
Project Partner in Ballast Water Opportunity (BWO), a North Sea INTER-REG project focusing on risk management of ballast water. Duration: 2009-2012. Budget, DHI: 100.000 €.
63
Year Firm Position and responsibilities
Project Manager of “Udvikling/tilvejebringelse af marine data til imple-mentering af Havstrategidirektivet i Østersø- og Nordsøregionen”, a project funded by the Danish Spatial and Environmental Planning Agency. The project is directly related to the BWO project. Duration: 2009-2011. Budget: 750.000 dkk.
Project Partner in EUSeaMap, a DG MARE funded project aiming to devel-
op broad-scale habitat maps for the North Sea, Baltic Sea and western Mediterranean Sea. Duration: 2009-2012. Budget, DHI: ~ 72.000 €.
Ecosystem-based management of eutrophication: Linking monitoring,
assessments, and the society. Internal project aiming at a Ph.D. degree. Co-funded by DHI via the RK contact 2006-2009 from the Danish Ministry of Science, Technology and Innovation. Duration: 2009-2010. Budget: 475.000 dkk.
Project Partner in an EU WFD Support Contract. Funded by the European
Commission (DG ENV). Focus is on compliance checking of European WFD River Basin Management Plans. Duration: 2009-2012. Total budget: 1.0 mio. €.
Contributor to several marine tasks under the EEA Topic Centre on Water (ETC/Water). Duration: 2008-2010. In 2010, focus is on tasks related to marine protected areas, assessment tools as well as pan-European marine assessments (SoE 2010). Annual budget for ETC/Water: ~ 1.150.000 €. Annual budget for DHI in 2007-2009: ~ 100.000 €. Budget for 2010 is ~ 45.000 €.
Project Manager of KARMA, a project funded by the Danish Spatial and Environmental Planning Agency and the Swedish EPA focusing on delinea-tion of and data availability within the Kattegat and its catchment area. Oth-er partners are the National Environmental Research Institute and Swedish Meteorological and Hydrological Institute. Duration: 2009-2010. Budget: 55.000 €, DHI’s share is 40%.
Conference Secretary for the 3rd International Symposium on Research
and Management of Eutrophication in Coastal Ecosystems (EUTRO 2010), which takes place 15-18 June 2010. Duration: 2009-2011. Please see www.eutro2010.dhi.dk for details. Budget, phase I: 300.000 dkk. Budget, phase II: ~ 2.0 mio. dkk.
Project Partner in HELCOM HOLAS: The HELCOM Holistic Assessment of the Baltic Sea Environment. Chair of the HELCOM HOLAS Task Force as well as contributor. Funded by DG Environment and the Ministry of Envi-ronment in Sweden. Duration: 2009-2010. Budget, DHI: 79.600 €.
Member of the EU Task Group on Eutrophication (MSFD ETG). Funded by the Danish Spatial and Environmental Planning Agency. Duration: 2009. Budget: 125.000 dkk.
Guest editor of a Special Issue of Hydrobiologia (together with Prof. Daniel J. Conley, Lund University, Sweden). The Special Issue includes 22 papers based on presentation from the International Symposium on Re-search and Management of Eutrophication in Coastal Ecosystems (EUTRO 2006), which took place in June 2006.
64
Year Firm Position and responsibilities
Project Manager of MARCOS: A NMR funded project focusing on “Marine European Directives: Concepts, Overlap and Synergies”. Duration: 2007-2009. Budget: 725.000 dkk.
Project Supervisor of CONFIRM: A NMR funded project focusing on confi-
dence rating of eutrophication assessments. Duration 2008-2009. Budget: 300.000 dkk.
Work Package Leader (WP6) in HELCOM BIO, phase II: An integrated thematic assessment of biodiversity in the Baltic Sea. The focus is on de-velopment of a tool for assessment of conservation status in the Baltic Sea. Duration: 2007-2008. Budget is linked to EUTRO-PRO and “Udvikling af et marint tilstandsvurderingsværktøj for Natura 2000 områder”.
Project Partner in MOPODECO: A NMR funded project focusing on model-
ling of habitats in the Baltic Sea. Duration: 2008-2009. Budget: 1.2 mio. dkk.
Project Manager of OxyBas, phase I: A project aiming at a Full Application to the Swedish EPA / FORMAS describing the project “Oxygenation of sed-iments in the Baltic Sea by ecological engineering” (OxyBas, phase II). Du-ration: 2008. Budget: 100.000 sek.
Project Manager of BALANCE Synthesis, part 1 and 2: A follow-up project on the BALANCE project, funded by the Danish Spatial and Environmental Planning Agency. Duration: 2008-2009. Budget: 300.000 dkk.
Project Manager of HELCOM EUTRO-PRO: An integrated thematic as-sessment of eutrophication in the Baltic Sea. Duration: 2006-2008. Budget for 2006/2007: 27.000 €. Budget for 2007/2008: 30.000 €.
Project Manager of BALANCE, an INTERREG IIIB project focusing on mapping and management of marine habitats as well as development of templates and tools for marine spatial planning in the Baltic Sea. Duration: 2005-2007. Budget: 4.700.000 €. Budget related to spin-out projects: 1.2 mio. dkk.
Project Supervisor of “Fagligt grundlag for genetablering af stenrev i Limfjorden”: A pilot project funded by the Danish Forest & Nature Agency focusing on restoration of stone reefs in Limfjorden as well as development of ‘supplementary measures’ sensu the WFD. Duration: 2007-2008. Budget for pilot project: 500.000 dkk.
Project Partner in IGLOO: A NOVANA funded project developing climate change indicators. Duration: 2007-2008. Budget for DHI: 42.000 dkk.
Project Partner in “Udvikling af et marint tilstandsvurderingsværktøj for Natura 2000 områder”, sub-contracted by the National Environmental Re-search Institute: The project is funded by the Danish Spatial and Environ-mental Planning Agency focusing on developing a tool for assessment of ‘conservation status’ sensu the EC Habitats Directive. Duration: 2007-2008. Budget for DHI: 52.000 dkk.
65
Year Firm Position and responsibilities
Marine Team Leader: European Topic Centre on Water (ETC/Water). Duration: 2007. Annual budget for ETC/Water: ~1.045.000 €.
Project Supervisor of OSPAR COMP-2: A project funded by the Danish
EPA focusing on assessment of eutrophication in the North Sea, Skagerrak and Kattegat. Duration: 2007-2008. Budget: 300.000 dkk.
Project Manager of HELCOM BIO, phase I: A pilot project producing a
synopsis for the production of an integrated thematic assessment on biodi-versity in the Baltic Sea. Duration: 2006. Budget: 20.000 €.
Project Manager of BSPC EUTRO: Writing a booklet on eutrophication in
the Baltic Sea focusing on effects, causes and solutions. Duration: 2006. Budget: 100.000 dkk.
Project Manager of MST CO-EUTRO. Duration: 2005-2007. Budget: 300.000 dkk.
Conference Secretary for the 2nd International Symposium on Research
and Management of Eutrophication in Coastal Ecosystems (EUTRO 2006), which took place 20-23 June 2006. Duration: 2005-2006. Budget: 2.8 mio. dkk.
Project Manager of MST HEAT, a project supporting HELCOM EUTRO.
Duration: 2005. Budget: 200.000 dkk.
Project Manager of HELCOM EUTRO (Development of tools for a thematic
eutrophication assessment). Duration: 2005. Budget: 30.000 €.
Project Manager of SNS/MST SYNERGY (Synergies and overlap between the EC Habitats Directive and the EU Water Framework Directive). Dura-tion: 2005. Budget: 100.000 dkk.
2001 – 2004 National Environmental Research Institute, Dept. of Marine Ecology
Chief Consultant and Head of the Danish National Marine Monitoring Focal Point (M-FDC), secretary of the National Steering Group on Marine Monitor-ing and Assessment.
Co-ordination of the revision of the NOVA programme (1998-2003) and the design of the NOVANA programme (2004-2009).
Co-ordinator of the departments work in relation to HELCOM and OSPAR plus participation in relevant committees and working groups working with monitoring and assessment of environment and nature in Danish waters.
Danish head-of-delegation in the OSPAR Assessment and Monitoring Group (ASMO). Danish marine representative in Nordic Council of Minis-ters’ Sea and Air Group, and Danish representative in the HELCOM/ICES SGQAB.
Participant in the WFD CIS Coast group and the pan-European Eutrophica-tion Activity.
Participation in EU projects under the 5th framework programme (FP5): CHARM, DANLIM and MOLTEN.
66
Year Firm Position and responsibilities 1999 – 2000
Danish Environmental Protection Agency, Waste Water and Groundwater Division. (Until May 2000: Waste Water and Aquatic Monitoring Division)
Implementation of the Water Framework Directive, in particular regarding principles for establishing ecological quality standards.
In addition, (i) political/ administrative summary report on status and per-spectives for management of the Danish aquatic environment (Vandmiljø-2000), (ii) co-ordination of the Agency’s viewpoints in relation to the Nation-al Monitoring Board and NOVA-2003, and (iii) National Steering Group on Hydrological Point Sources.
Co-editor of the NOVA-2003 programme document, Vandmiljø-99, national guidelines on reporting of the NOVA programme, guidelines on annual evaluation of sampling, data flow and reporting.
Co-ordination of Danish participation in the HELCOM MONAS group.
1999 National Forest and Nature Agency, Ecological Division (ED)
Legal and technical casework in relation to the Danish Watercourse Act.
Supervision in relation to the surveillance monitoring of freshwaters carried of by the Danish counties
ED Work Programme 1999 and RTD projects.
1998 Danish Environmental Protection Agency, Freshwater and Waste Water Division
Casework in relation to the Danish Watercourse Act.
Evaluation report on the 1993-1997 National Monitoring Programme.
Guidelines for 1999 reporting of the NOVA programme.
A synopsis for HELCOM’s 4th Periodic Assessment of the State of the Marine Environment of the Baltic Sea 1994-1998 on behalf of the Marine Division.
1994 – 1998 Danish Environmental Protection Agency,
Marine Division Participation in the preparation of the 4th North Sea Conference in 1995.
Marine fish farming 1994-1998 (technical casework, data management and reporting).
Danish Head-of-Delegation HELCOM EC 1994-1998.
Professional Secretary to the National Monitoring Board (1995-1998) and the National Revision Task Team (1996-1998).
Chairman of the National Steering Group on Marine Monitoring (1995-1998).
Preparation of a large number of meetings and an equal number of sum-mary records.
Co-ordination of the ’closing down’ of the Danish Marine Research Pro-gramme (Hav90), and contribution of the report summarising the most im-portant results of Hav90.
1992 – 1994 BioConsult (now SBH-consult)
Consultant for the Danish EPA. Task: management of the Danish Marine Re-search Programme (Hav90). The work included general project management, including reimbursement, budgets, publication of project reports, secretary to the National Marine Research Advisory Board well as planning of the International Symposium on Nutrient Dynamics in Coastal and Estuarine Environments, 13-16 October 1993 (EUTRO 1993), the latter including work in relation to the produc-tion of Symposium Proceedings.
1990 – 1992 Nordic Council of Ministers (Danish Envi-ronmental Protection Agency)
Project secretary working for the Nordic Council of Ministers (NCM) project on marine monitoring. The main objective of the project was to compile a framework for an improved co-ordination of marine monitoring in the Nordic countries. Con-clusions and recommendations from this work have to a large extent influenced the Baltic Monitoring Programme/ COMBINE and the marine sub-programme of NOVA-2003.
1988 – 1990 National Environmental Research Institute, Dep. of Marine Ecology
Student assistant in M-FDC and for some periods working with research projects (NPo Research Programme and Hav90).
67
Languages Danish English German French Swedish
Speaking 5 5 2 1 4
Reading 5 5 2 2 5
Writing 5 5 1 1 3
(Mother tongue/excellent: 5; Average: 3-4; Poor: 1-2)
Publications in English
Currently, I have more than 80 publications in English. Most are Technical Reports, but the list also in-cludes a growing number of peer reviewed papers as well as three reviewed books and a special issue of a scientific journal. As a spin off, I have reviewed manuscripts for the following scientific journals: (1) Biogeochemistry; (2) Environmental Management; (3) Environment International; (4) Estuaries & Coasts (formerly Estuaries); (5) Estuarine, Coastal, and Shelf Science; (6) Hydrobiologia; and (7) Marine Biology Research.
= indicates a reviewed publication. Bold = indicates an assessment.
82 2012 Laamanen, M., S. Korpinen, U.L. Zweifel & J.H. Andersen (submitted): Ecosystem health In: Biological Oceanography of
the Baltic Sea (Eds: Pauline Snoeijs, Hendrik Schubert & Teresa Radziejewska). Springer. 81 Andersen J.H. (accepted): Ecosystem-based management of coastal eutrophication. Connecting science, policy and socie-
ty. Ph.D. thesis. University of Copenhagen. 56 pp + annexes. 80 Andersen, J.H. (accepted): BEAT, the HELCOM Biodiversity Assessment Tool. In: Skov et al. (accepted): Modelling of the
potential coverage of habitat-forming species and development of tools to evaluate the conservation status of the ma-rine Annex I habitats. Tema Nord Report. Nordic Council of Ministers.
79 Carstensen, J., J.H. Andersen, K. Dromph, V. Fleming-Lehtinen, S. Simis, B. Gustavsson, A. Norkko, H. Radke, D.L.J. Petersen & T. Uhrenholdt (accepted): Approaches and methods for eutrophication target setting in the Baltic Sea re-gion. Baltic Sea Environemt Proceesings. 133 pp.
78 Andersen, J,H., J.W. Hansen, M. Mannerla, S. Korpinen & J. Reker (accepted): A glossary of terms commonly used in the Marine Strategy Framework Directive. NERI Technical Report. 31 pp.
77 Korpinen, S., L. Meski, J.H. Andersen & M. Laamanen (2012): Human pressures and their potential impact on the Baltic Sea ecosystem. Ecological Indicators 15:105-114. http://dx.doi.org/10.1016/j.ecolind.2011.09.023
76 2011 Andersen, J.H., P. Axe, H. Backer, J. Carstensen, U. Claussen, V. Fleming-Lehtinen, M. Järvinen, H. Kaartokallio, S. Knuuttila, S. Korpinen, M. Laamanen, E. Lysiak-Pastuszak, G. Martin, F. Møhlenberg, C. Murray, G. Nausch, A. Norkko, & A. Villnäs (2010): Getting the measure of eutrophication in the Baltic Sea: towards improved assessment principles and methods. Biogeochemistry. DOI 10.1007/s10533-010-9508-4. http://www.springerlink.com/content/x76wq76863458471/fulltext.pdf
75 Andersen, J.H. & J. Carstensen (2011): Reference conditions and acceptable deviation: Concepts, definitions and their practical use. HELCOM TARGREV Working Document. 21 pp.
74 Ferreira, J.G., J.H. Andersen, A. Borja, S.B. Bricker, J. Camp, M. Cardoso da Silva, E. Garcés, A.-S. Heiskanen, C. Hum-borg, L. Ignatiades, C. Lancelot, A. Menesguen, P. Tett, N. Hoepffner & U. Claussen (2011): Indicators of human-in-duced eutrophication to assess the environmental status within the European Marine Strategy Framework Directive. Estuarine, Coastal and Shelf Science. DOI: 10.1016/j.ecss.2011.03.014.4
73 Murray, C., J.H. Andersen, H. Kaartokallio, P. Axe, J. Molvær, K. Norling & M. Krüger-Johansen (2011): Confidence rating of marine eutrophication assessments. Tema Nord 2011:504. 75 pp.
72 Andersen, J.H., S. Bricker, J. Carstensen & J.E. Larsen (2011): ICES/DHI/NOAA Third International Symposium on Re-search and Management of Eutrophication in Coastal Ecosystems. 6 pp + Online Supplementary Material. In: ICES (2011): ICES Symposium Report 2010. ICES CM 2011/GEN 0x.
4 This paper was highlighted by the European Commission in ’Science for Environment Policy’, a news alert service from DG
Environment, please see: http://ec.europa.eu/environment/integration/research/newsalert/pdf/252na4.pdf.
68
71 2010 ETC/W (2010): Delineation of Marine Regions in EU Directives and Regional Sea Conventions. Draft Scoping Report edited by A. Stock, J.H. Andersen, B.M. Sharry & H.M. Jensen. 22 pp.
70 EUSeaMap (2010): EUSeaMap Final Report. Preparatory Action for development and assessment of a European broad-scale seabed habitat map. Contribution by J. Reker & J.H. Andersen. EC contract no. MARE/2008/07. 223 pp.
69 Kaartokallio, H., J.H. Andersen, J.N. Jensen, A. Künitzer, N. Green & M. Peterlin (2010): Review of precursors on initial assessment from Regional Marine Conventions. Scoping report from ETC/W to EEA. 28 pp.
68 Korpinen, S, L. Meski, J.H. Andersen & M. Laamanen (2010): Towards a tool for quantifying anthropogenic pressures and potential impacts on the Baltic Sea marine environment. A background document on the method, data and testing of the Baltic Sea Pressure and Impact indices. Baltic Sea Environmental Proceedings No. 125. 73 pp.
67 Christiansen, T. (Ed.), A. Meiner, B. Werner, C. Romao, E. Gelabert, R.P. Collins, R. Uhel, A. Ruus, A.-S. Heiskanen, A. Künitzer, A. Raike, B. Bjerkeng, C. Emblow, G. Coppini, H. Sparholt, J.H. Andersen, J.-M. Leppänen, J.N. Jensen, M. Lago, M. Peterlin, N. Pinardi, N. Holdsworth, N. Green, P. Degnbol, B. Mac Sharry, S. Condé, A.I. Campos, J. Orr & S. van den Hove (2010): The European Environment. State and Outlook 2010. Marine and Coastal Environment. European Environment Agency, Copenhagen. 58 pp.
66 J.H. Andersen & S.B. Bricker (2010): EUTRO 2010 Report-out. CERF Newsletter 36(3):24. 65 DHI, ICES & NOAA (2010): Third International Symposium on Research and Management of Eutrophication in Coastal Eco-
systems. 15-18 June 2010, Nyborg, Denmark. Programme and Abstracts. Edited by J.H. Andersen. 44 pp. 64 Andersen, J.H., L. Hasselström, S. Korpinen, M. Laamanen, A. Soutukorva & U. Volpers (2010): Chapter 1: Introduc-
tion. Pages 6-13 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Environmental Proceedings 122. Helsinki Commission. 63 pp.
63 Andersen, J.H., S. Korpinen, M. Laamanen & C. Murray (2010): 2.1 Integrated and Holistic Assessments. Pages 14-15 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Environmental Proceedings 122. Helsinki Commission. 63 pp.
62 Andersen, J.H. (2010): 2.2 Eutrophication. Pages 16-17 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Environmental Proceedings 122. Helsinki Com-mission. 63 pp.
61 S. Korpinen, J.H. Andersen, M. Laamanen & C. Murray (2010): 2.3 Hazardous substances. Pages 18-21 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Envi-ronmental Proceedings 122. Helsinki Commission. 63 pp.
60 Zweifel, U.L., S. Korpinen, R. Ljungberg & J.H. Andersen (2010): 2.4 Biodiversity. Pages 22-26 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Environmental Proceedings 122. Helsinki Commission. 63 pp.
59 M. Laamanen, J.H. Andersen, U. Claussen, M. Durkin, S. Korpinen, J. Reker, M. Stankiewicz & U. Volpers (2010): Chapter 4: What are the solutions? Pages 42-49 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Baltic Sea Environmental Proceedings 122. Helsinki Com-mission. 63 pp.
58 U. Volpers, Andersen, J.H., S. Korpinen & M. Laamanen (2010): Chapter 6: Conclusions and outlook. Pages 54-57 in: HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Bal-tic Sea Environmental Proceedings 122. Helsinki Commission. 63 pp.
57 HELCOM (2010): Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment 2003-2007. Edited by J.H. Andersen, S. Korpinen, M. Laamanen & U. Wolpers. Baltic Sea Environmental Proceedings 122. Helsinki Commission. 63 pp. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep122.pdf
56 HELCOM (2010): Hazardous substances in the Baltic Sea. An integrated thematic assessment of hazardous sub-stances in the Baltic Sea. Edited by S. Korpinen & M. Laamanen with contributions from J.H. Andersen, L. Asplund, U. Berger, A. Bignert, E. Boalt, K. Broeg, A. Brzozowska, I. Cato, M. Durkin, G. Garnaga, K. Gus-tavson, M. Haarich, B. Hedlund, P. Köngäs, T. Lang, M.M. Larsen, K. Lehtonen, J. Mannio, J. Mehtonen, C. Murray, S. Nielsen, B. Nyström, K. Pazdro, P. Ringeltaube, D. Schiedek, R. Schneider, M. Stankiewicz, J. Strand, B. Sundelin, M. Söderström, H. Vallius, P. Vanninen, M. Verta, N. Vieno, P. Vuorinen and A. Zaharov. Baltic Sea Environmental Proceedings 120B. Helsinki Commission. 116 pp.
55 Andersen, J.H., C. Murray, H. Kaartokallio, P. Axe & J. Molvær (2010): A simple method for confidence rating of eutrophi-cation status assessments. Marine Pollution Bulletin 60:919-924. doi:10.1016/j.marpolbul.2010.03.020. http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6V6N-4YT7KN7-9-7&_cdi=5819&_user=684530&_pii=S0025326X10001104&_orig=search&_coverDate=06%2F30%2F2010&_sk=999399993&view=c&wchp=dGLbVlb-zSkzk&md5=ad2b416d4eff4bf7e9a3a99eb6bbb9aa&ie=/sdarticle.pdf
54 Andersen, J.H., J. Dørge, H. Skov, A. Stock, J. Carstensen, K. Dahl, M. Hjorth, A.B. Josefsson, M.M. Larsen, J. Strand, P. Andersson, P. Axe, J. Reker & S. Korpinen (2010): Delineation scenarios for the Kattegat, data availability and mana-gement support tools. DHI Technical Report to the Agency for Spatial and Environmental Planning, Denmark. 86 pp.
53 ETC/W (2010): Assessment of the European Marine Environment. Background Report for the SoER 2010, part B. By: J.-M. Leppänen, A.-S. Heiskanen, M. Viitasalo, H. Rouse (Eds.), M. Peterlin, C. Emblow, N.W. Green, G. Coppini, J. Dorandeu, G. Larnicol, S. Marullo, P. Lowe, N. Pinardi, J.N. Jensen, J.H. Andersen, H. Peltonen, M. Raateoja & A. Räike. European Environment Agency. 111 pp.
52 Ferreira, J.G., J.H. Andersen, A. Borja, S.B. Bricker, J. Camp, M. Cardoso da Silva, E. Garcés, A.-S. Heiskanen, C. Hum-borg, L. Ignatiades, C. Lancelot, A. Menesguen, P. Tett, N. Hoepffner & U. Claussen (2010): Marine Strategy Framework Directive. Task Group 5 Report. Eutrophication 49 pp.
69
51 2009 Andersen, J.H. & D.J. Conley (editors) (2009): Eutrophication in Coastal Ecosystems. Towards better understanding and management strategies. Developments in Hydrobiology 207. 269 pp. Previously published in Hydrobiologia 629(1). http://www.springer.com/environment/aquatic+sciences/book/978-90-481-3384-0.
50 Laamanen, M. & J.H. Andersen (2009): Eutrophication. In: HELCOM (2009): Biodiversity in the Baltic Sea. An integrated thematic assessment of biodiversity and nature conservation in the Baltic Sea. Ed. by U.L. Zweifel. Baltic Sea Environmental Proceedings No. 116B. Helsinki Commission. 188 pp.
49 Backer, H. & J.H. Andersen (2009): Towards an indicator-based assessment of the Baltic Sea biodiversity. In: HELCOM (2009): Biodiversity in the Baltic Sea. An integrated thematic assessment of biodiversity and nature conservation in the Baltic Sea. Ed. by U.L. Zweifel. Baltic Sea Environmental Proceedings No. 116b. Helsinki Commission. 188 pp.
48 Andersen, J.H. & D.J. Conley (guest editors) (2009): Eutrophication in coastal ecosystems. Hydrobiologia. 629(1), 269 pp. http://www.springerlink.com/content/t8j727n2k266/?p=1b40a1097a1b47a9842b02ba104ea0b6&pi=5
47 Andersen, J.H. & D.J. Conley (2009): Eutrophication in coastal marine ecosystems: towards better understanding and management strategies. Hydrobiologia 629(1):1-4. http://www.springerlink.com/content/w707717n84j65571/fulltext.pdf
46 Andersen, J.H., S. Korpinen & M. Laamanen (2009): Towards a holistic assessment of environmental status in the Baltic Sea. HOLAS roadmap. DHI Technical Report to HELCOM. 43 pp.
45 Dahllöf, I. & J.H. Andersen (2009): Hazardous and Radioactive Substances in Danish Marine Waters. Status and Temporal Trends. Danish Spatial and Environmental Planning Agency & National Environmental Research Institute. 110 pp. http://www2.dmu.dk/pub/OSPAR_Hazardous_Substances_print.pdf
44 HELCOM (2009): Eutrophication in the Baltic Sea. An integrated thematic assessment of eutrophication in the Baltic Sea region: Executive Summary. Edited by J.F. Pawlak, M. Laamanen & J.H. Andersen. Baltic Sea Environmental Proceedings No. 115A. Helsinki Commission. 19 pp.
43 HELCOM (2009): Eutrophication in the Baltic Sea. An integrated thematic assessment of eutrophication in the Baltic Sea region. Ed. by J.H. Andersen & M. Laamanen. Baltic Sea Environmental Proceedings No. 115B. Helsinki Commission. 148 pp.
http://meeting.helcom.fi/c/document_library/get_file?p_l_id=79889&folderId=377779&name=DLFE-36818.pdf 42 2008 Andersen, J.H. & H. Backer (2008): Development of an indicator-based tool for assessment of biodiversity in the Baltic
Sea. DHI Technical Report to HELCOM HABITAT and HELCOM BIO. 34 pp. 41 Andersen, J.H. & H. Kaas (2008): Danish assessment of eutrophication status in the North Sea, Skagerrak and
Kattegat: OSPAR Common Procedure 2001-2005. DHI Technical Report to the Danish Spatial and Environmental Planning Agency. 86 pp.
40 ETC/Water (2008): Improving EEA marine indicators. A review of their performance and suggested ‘next steps’. Final Draft Scoping Report edited by J.H. Andersen, P. Kuuppo, T. Christiansen & E.R. Gelabert. 77 pp.
39 2007 Andersen, J.H., A. Erichsen, K. Garde, C. Murray & F. Møhlenberg (2007): Strengthening the Tools for Assessment of Coastal Eutrophication in Russian Waters of the Baltic Sea. DHI Technical Report to the Danish Environmental Protection Agency. 61 pp.
38 Hansen, I.S., N. Keul, J.T. Sørensen, A. Erichsen & J.H. Andersen (2007): Baltic Sea oxygen maps. BALANCE Interim Report No. 17. 36 pp.
37 Andersen, J.H. & F. Møhlenberg (2007): Testing of the draft HELCOM Eutrophication Assessment Tool (HEAT) in 45 basins and coastal water bodies of the Baltic Sea. DHI Technical Report to HELCOM. 50 pp.
36 Andersen, J.H., & A. Erichsen (2007): Modelling of reference conditions in the Baltic Sea. DHI Technical Report to HELCOM. 18 pp. + annexes.
35 Al-Hamdani, Z. & J. Reker (eds.), J.H. Andersen and 22 others (2007): Towards marine landscapes in the Baltic Sea ecoregion. BALANCE Interim Report No. 10. 117 pp.
34 EEA & ETC/Water (2007): Towards a ‘converging’ framework for marine monitoring and assessment of European marine waters. Synthesis of EEA-led workshops on Operational oceanography, Ecological processes and biological elements and Chemical loads and burdens. Edited by E.R. Gelabert & J.H. Andersen. 43 pp. + annexes.
33 Andersen, J.H & J. Reker (2007): BALANCE Newsletter No. 3. 4 pp. 32 Andersen, J.H., C. Murray & H. Skov (2007): Interim overview of reporting obligations, monitoring activities and data
available for the HELCOM integrated thematic assessment of biodiversity and nature conservation in the Baltic Sea. DHI Technical Report to HELCOM. 29 pp.
31 2006 Andersen, J.H., H.B. Nielsen & H. Skov (2006): Getting the measure of biodiversity in the Baltic Sea. DHI Technical Report to HELCOM. 21 pp.
30 HELCOM (2006): Development of tools for assessment of eutrophication in the Baltic Sea. Baltic Sea Environment Pro-ceedings 104. 62 pp. Edited by J.H. Andersen.
29 Andersen, J.H. (2006): Project Proposal: Towards an integrated thematic assessment of eutrophication in the Baltic Sea. DHI Technical Report to HELCOM. 46 pp.
28 DHI, Danish EPA, Fyn County & Swedish EPA (2006): Research and Management of Eutrophication in Coastal Ecosys-tems. An International Symposium, 20-23 June 2006, Nyborg, Denmark. Programme and Book-of-Abstracts. 94 pp. Edited by J.H. Andersen.
27 Andersen, J.H & J. Reker (2006): BALANCE Newsletter No. 2. 4 pp. 26 Andersen, J.H & J.T. Pawlak (2006): Nutrients and Eutrophication in the Baltic Sea. Effects / Causes / Solutions. Booklet
produced for the Baltic Sea Parliamentary Conference. 32 pp. 25 Andersen, J.H., L. Schlüter & G. Ærtebjerg (2006): Coastal eutrophication: recent developments in definitions and implica-
tions for monitoring strategies. Journ. Plankt. Res. 28(7):621-628. http://plankt.oxfordjournals.org/cgi/reprint/28/7/621
70
24 Petersen, J.K., J.H. Andersen, K. Dahl, O.S. Hansen, A.B. Josefson, J. Karlsson, L.-O. Loo, J. Magnusson, F. Moy & P. Nilsson (2006): Reference conditions and EQOs for aquatic vegetation and macrozoobenthos. TemaNord 2006:510. 138 pp.
23 Carstensen, J., D.J. Conley, J.H. Andersen & G. Ærtebjerg (2006): Coastal eutrophication and trend reversal: A Danish case study. Limnology & Oceanology 51:398-408. http://aslo.org/lo/toc/vol_51/issue_1_part_2/0398.pdf
22 2005 Andersen, J.H. & J. Reker (2005): BALANCE Newsletter No. 1. 4 pp. 21 Petersen, J.K., Hansen, O.S., Henriksen, P., Carstensen, J., Krause-Jensen, D., Dahl, K., Josefson, A.B., Hansen, J.L.S.,
Middelboe, A.L. & Andersen, J.H. (2005): Scientific and technical background for intercalibration of Danish coastal waters. National Environmental Research Institute, Denmark. 72 pp. - NERI Technical Report No. 563.
20 Andersen, J.H, F. Møhlenberg, T. Uhrenholdt, M.H. Jensen, B. Sømod & P. Henriksen (2005): Testing of the HELCOM Eutrophication Assessment Tool (HEAT) in Danish Marine Waters. DHI Technical Report. 32 pp.
19 Andersen, J. H. (Ed.) (2005) Marine waters. In: Svendsen, L. M., Bijl, L. van der, Boutrup, S. and Norup, B. (Eds.) 2005: NOVANA: Nationwide Monitoring and Assessment Programme for the Aquatic and Terrestrial Environments. Pro-gramme Description - Part 2. National Environmental Research Institute, Denmark. 137 pp. - NERI Technical Report No. 537.
18 2004 Andersen, J.H., D.J. Conley & S. Hedal (2004): Palaeo-ecology, reference conditions and classification of ecological status: The EU Water Framework Directive in practice. Mar. Poll. Bul. 49:282-290. http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6V6N-4CPDF84-3-7&_cdi=5819&_user=684530&_orig=search&_coverDate=08%2F31%2F2004&_sk=999509995&view=c&wchp=dGLbVlW-zSkWb&md5=a3554a13dd8d91c63139f6cc2593fc31&ie=/sdarticle.pdf
17 Christiansen, T., J. Andersen & J.B. Jensen (2004): Defining a typology for Danish coastal waters. Coastline Reports 4(2004):49-54.
16 Dahl, K., Larsen, M.M., Rasmussen, M.B., Andersen, J.H., Petersen, J.K., Josefson, A.B., Lundsteen, S., Dahllöf, I., Chris-tiansen, T., Helmig, S.A. & Reker, J. (2004): Tools to assess the conservation status of marine habitats in special are-as of conservation. Phase 1: Identification of potential indicators and available data. National Environmental Research Institute. – Technical Report from NERI.
15 2003 Andersen, J.H. & O.S. Hansen (2003): Background, definition, causes and effects. Pages 7-17 in: Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Nutrients and Eutrophication in Danish Marine Waters. A Challenge to Sci-ence and Management. National Environmental Research Institute. 126 p.
14 13
Andersen, J.H. & O.S. Hansen (2003): Fish kills in coastal waters. Pages 80-83 in: Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Nutrients and Eutrophication in Danish Marine Waters. A Challenge to Science and Man-agement. National Environmental Research Institute. 126 p.
Andersen, J.H., J.B. Jensen & H. Karup (2003): Responses and adaptive management. Pages 85-99 in: Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Nutrients and Eutrophication in Danish Marine Waters. A Challenge to Science and Management. National Environmental Research Institute. 126 p.
12 Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Summary, conclusions and the future. Pages 101-109 in: Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Nutrients and Eutrophication in Danish Marine Waters. A Challenge to Science and Management. National Environmental Research Institute. 126 p.
11 Ærtebjerg, G., J.H. Andersen & O.S. Hansen (2003): Nutrients and Eutrophication in Danish Marine Waters. A Chal-lenge to Science and Management. National Environmental Research Institute. 126 p. http://www2.dmu.dk/1_Viden/2_Publikationer/3_ovrige/rapporter/Nedmw2003_alle.pdf
10 2002 Conley, D., S. Markager, J. Andersen, T. Ellermann & L.M. Svendsen (2002): Coastal Eutrophication and the Danish National Aquatic Monitoring and Assessment Program. Estuaries 25(4b): 848-861. http://www.springerlink.com/content/95780457274837t6/fulltext.pdf
9 2001 Conley, D., J. Andersen, J. Carstensen & P. Henriksen (2001): Long-term trends in nutrient loading, nutrient concentrations and nutrient limitation in Danish estuaries. OSPAR MON 2001.
8 2000 Aquatic Environment 1999. State of the Danish Aquatic Environment. Environmental Investigations, no. 3/2000. Eds: J. Andersen & D. Barry. 138 p.
7 1998 Christensen, P.B., F. Møhlenberg, L.C. Lund-Hansen, J. Borum, C. Christiansen, S.E. Larsen, M.E. Hansen, J. Andersen & J. Kirkegaard (1998): The Danish Marine Environment: Has Action Improved its State? - Havforskning fra Miljøstyrel-sen, nr. 62. 115 p.
6
Andersen, J. (1998): Draft Synopsis for the 4th Periodic Assessment of the State of the Marine Environment of the Baltic Sea 1994-1998. Version 1.0. 28 p + annexes.
5 1996 Aquatic Environment 1994. Overall Trend in Point-source Discharges and Status of the Danish Aquatic Environ-ment. Environmental Investigations, no. 1/1996. Eds: J. Andersen & T. Christensen. 151 p.
4 Scientific Symposium on the North Sea Quality Status Report, 18-21 April 1994, Ebeltoft, Denmark. Proceedings. Eds.: J. Andersen, H. Karup & U.B. Nielsen. Danish Environmental Protection Agency 1996. 346 p.
3 1995 Progress Report. 4th International Conference on the Protection of the North Sea. Eds.: J. Andersen & T. Niilonen. Danish Environmental Protection Agency, 1995. 247 p.
2 1992 Andersen, J. & J. Kirkegaard (1992): The Marine Research Programme in Denmark. North Sea Task Force News 4:6-7. 2 p.
1 1991 A Nordic Strategy for the Co-ordination and Enhancement of Marine Monitoring and Assessment. Ed.: J. Andersen. HEL-COM EC2/INF.7, 1991. 8 p.
71
Publications in Danish
= indikerer at publikationen er reviewet. Fed = indikerer at publikationen er en tilstandsrapport/assessment.
57 2012 Bach, H., I.K. Petersen, R.D. Nielsen, T. Fox, C. Topping, B. Nygaard, M. Elmeros, B. Søgaard, J. Kahlert, S. Sveegaard, R.
Dietz, J. Tougaard, J.N. Nielsen, J. Teilmann, A.B. Josefson, C. Mohn, J.L.S. Hansen, K. Timmermann, J.H. Andersen, M. Maar, H.H. Jacobsen, E. Friis Møller, K. Dahl, L.C. Lund-Hansen, P. Grønkjær, M.B. Rasmussen, M. Winther, L. Mar-tinsen, M. Zandersen, H.R. Olesen & B. Münier (accepted): Miljøfagligt review af VVM-redegørelsen for Femern forb-indelsen. Miljø-beskrivelsen. Videnskabelig rapport fra DCE - Nationalt Center for Miljø og Energi. 86 pp.
56 Andersen, J.H. & C. Murray (accepted): Foreløbig integreret vurdering af miljøtilstanden i de danske farvande - en indikator-baseret vurdering. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 11 pp.
55 Andersen, J.H., M.M. Larsen, C. Murray & J. Strand (accepted): En integreret vurdering og klassifikation af den kemiske til-stand i de danske farvande - en indikatorbaseret vurdering. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 11 pp.
54 Andersen, J.H., S. Korpinen & A. Stock (accepted): Foreløbig vurdering af kumulative påvirkninger og belastninger i de dan-ske farvande. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 14 pp.
53 Hansen, J.W., J.H. Andersen, J. Strand & T.K. Sørensen (accepted): Affald i havet. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 28 pp.
52 J.H. Andersen, J.W. Hansen & J. Carstensen (accepted): Væsentlige ændringer i temperatur- og salinitetsforholdene i de danske farvande forårsaget af menneskelige aktiviteter. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 13 pp.
51 J.H. Andersen, C. Göke & C. Murray (accepted): Klassifikation af af biodiversitetstilstanden i de danske farvande – en indi-kator-baseret statusvurdering. Fagligt notat fra DCE - Nationale Center for Miljø og Energi. 30 pp.
50 J.H. Andersen, C.D. Pommer, J.W. Hansen & P. Dolmer (accepted): Foreløbig karakterisering af fysisk skader forårsaget af råstofindvinding og bundtrawling i de danske farvande. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 27 pp.
49 J.H. Andersen, J.W. Hansen, C. Murray, C. Göke & D.LJ. Petersen (accepted): Klassifikation af eutrofieringstilstanden i de danske farvande – en indikator-baseret statusvurdering. Fagligt notat fra DCE - Nationalt Center for Miljø og Energi. 42 pp.
48 2011 Andersen, J.H. & J. Carstensen (2011): Gisp. Vandmiljøplanerne virker. Politiken, 15. oktober 2011. Debat-sektionen side 8. 47 Andersen, J.H. (2011): ’Havet omkring Danmark’ – et forslag til synopsis for havstrategidirektivets basisanalyser. Notat fra
DMU. 49 pp. 46 2009 Josefson, A., D. Krause-Jensen, M.B. Rasmussen, J.H. Andersen & P. Henriksen (2009): Udvikling af indikatorer og tilstands-
vurderingsværktøj for marine Natura 2000 områder. Faglig rapport fra DMU, nr. 701. 76 pp. 45 2008 Møhlenberg, F., J.H. Andersen (Eds.), C. Murray, P.B. Christensen, T. Dalsgaard, H. Fossing & D. Krause-Jensen (2008):
Stenrev i Limfjorden: Fra naturgenopretning til supplerende virkemiddel. DHI Teknisk Rapport til By- og Landskabsstyrel-sen. 41 p + bilag.
44 Hansen, J.W., M. Nedergaard & F. Skov (Eds.) (2008): IGLOO – Indikatorer for globale klimaændringer i overvågningen. DHI rapport til Miljøcenter Ringkøbing, By- og Landskabsstyrelsen. 91 pp. Med bidrag fra J.H. Andersen.
43 Kaas, H. (Ed.) (2008): Ny teknologi i overvågningen. DHI Teknisk Rapport til By- og Landskabsstyrelsen. 96 pp. Med bidrag fra J.H. Andersen.
42 2005 Andersen, J.H. & H. Skov (2005): Synergi og overlap mellem Habitatdirektivet, Fuglebeskyttelsesdirektivet og Vandramme-direktivet – med fokus på kystvand. DHI Teknisk Rapport til Skov- og Naturstyrelsen og Miljøstyrelsen. 55 pp.
41 Andersen, J., S. Markager & G. Ærtebjerg (2005): Tekniske anvisninger for marin overvågning 2004 – 2009. Danmarks Miljø-undersøgelser. Kan downloades via: http://www.dmu.dk/Overvaagning/Fagdatacentre/Det+Marine+Fagdatacenter/Tekniske+anvisninger+NOVANA+2004-2009/
40 Andersen, J.H., Clarke, A., Conley, D.J., Dahllöf, I., Greve, T.M., Krause-Jensen, D., Larsen, M.M., Nielsen, K. & Reuss, N. (2005): Eksempler på økologisk klassificering af kystvande. Vandrammedirektiv-projekt fase IIIa. – Faglig rapport fra DMU nr. 530. 48 pp.
39 Svendsen, L.M., L. van der Bijl, S. Boutrup & B. Norup (2005): NOVANA. Det nationale program for overvågning af vandmiljøet og naturen. Programbeskrivelse - del 2. Faglig rapport fra Danmarks Miljøundersøgelser. 128 pp. Med bidrag af J.H. An-dersen.
38 Larsen, M.M., S. Foverskov & J.H. Andersen (2005): Havnesedimenter - Prøvetagning og analyser. Arbejdsrapport fra Miljø-styrelsen nr. 35, 2005. 77 pp.
37 2004 Ærtebjerg, G. & Andersen, J.H. (red.) og mange flere (2004): Marine områder 2003 – Miljøtilstand og udvikling. – Faglig rapport fra DMU nr. 513. 97 pp.
36 Andersen, J.H., J.B. Jensen, D. Krause-Jensen, H.B. Madsen & B. Riemann (2004): Fra vandmiljøplaner til vandplaner og indsatsprogrammer – med kvælstof som eksempel. 9 pp. I: Ærtebjerg & Andersen (red.) og mange flere (2004): Marine områder 2003 – Miljøtilstand og udvikling. - Faglig rapport fra DMU nr. 513. 97 pp.
72
35 Christensen, P.B, O.S. Hansen & G. Ærtebjerg (Eds.) (2004): Iltsvind. Med bidrag af J.H. Andersen, J. Carl, J. Carstensen, P. Clausen, R. Dietz, J. Fenger, T.M. Greve, J.L.S. Hansen, O. Hertel, A.B. Josefson, D. Krause-Jensen, A. Branth Peder-sen, I.K. Petersen, J.F. Steffensen & J. Teilmann. Forlaget Hovedland. 128 pp.
34 Ærtebjerg, G., Manscher O.H. & Andersen, J. (2004): Præliminær evaluering af marine data-værter i Sveriges marine over-vågningsprogram. Evalueringsnotat til Naturvårdsverkets Miljöövervakningsenhet. 19 pp.
33 Riemann, B. J.K. Petersen, D. Conley, A.B. Josefson, D. Krause-Jensen, J.H. Andersen, P. Henriksen, M.B. Rasmussen & P. Claussen (2004): Faglig udredning af problemer vedrørende tilstand og miljømål for Ringkøbing Fjord. Danmarks Miljø-undersøgelser, 51 pp.
32 2003 Fossing, H., Andersen, J.H. & Dalsgaard, T. (2003): Miljøtilpasset overvågning af saltvandsbaseret fiskeopdræt. Notat. 17 pp. 31 Rasmussen, M.B. & Andersen, J. (red.) og mange flere. (2003): Marine områder 2002 – Miljøtilstand og udvikling.
NOVA-2003. Danmarks Miljøundersøgelser. 98 pp. + bilag – Faglig rapport fra DMU nr. 467 30 Dahl, K., Larsen, M.M., Rasmussen, M.B., Andersen, J.H., Petersen, J.K., Josefsson, A.B., Lundsteen, S., Dahllöf, I. & Chri-
stiansen, T. (2003): Kvalitetsvurderingssystem for Habitatsdirektivets marine naturtyper. Fase I: Identifikation af potentiel-le indikatorer og tilgængelige data. Danmarks Miljøundersøgelser. 92 pp. – Faglig rapport fra DMU nr. 446.
29 Bendtsen, J., Andersen, J., Bendtsen, S.Å., Bruhn, B., Ellegaard, C., Rasmussen, J., & Vang, T. (2003): Kravspecifikation til dele af det marine modelkompleks. Oktober 2003. 27 pp.
28 Conley, D.J., A. Clarke, S. Juggins, F. Adser, N. Reuss & J. Andersen (2003): Vandrammedirektivet, næringsstoffer i kystvan-de (3). Vand & Jord 2/2003: 52-56. 5 pp.
27 2002 Ærtebjerg, G., Andersen, J., Carstensen, J., Christiansen, T., Dahl, K., Dahllöf, I., Fossing, H., Greve, T.M., Hansen, J.L.S., Henriksen, P., Josefson, A., Krause-Jensen, D., Larsen, M.M., Markager, S., Nielsen, T.G., Pedersen, B., Pe-tersen, J.K., Risgaard-Petersen, N., Rysgaard, S., Strand, J., Ovesen, N.B., Ellermann, T., Hertel, O., Skjøth, C.A. 2002: Marine områder 2001 - Miljøtilstand og udvikling. NOVA-2003. Danmarks Miljøundersøgelser. 94 pp. – Fag-lig rapport fra DMU nr. 419.
26 Pedersen, S., J. Andersen, J.G. Dannisøe, H. Kaas & F. Møhlenberg (2002): Vandrammedirektivet, konkretisering af miljømål (2). Vand & Jord 1/2002: 25-29, 5 pp.
25 2001 Andersen, J. (Red.) (2001): Fremtidens havovervågning. Revisionsscenarium med forslag til program for integreret overvåg-ning af miljø- og naturforhold i de danske farvande 2004 – 2009. 54 pp.
24 Henriksen, P, J. Andersen og mange flere (2001): Marine områder 2000. Miljøtilstand og udvikling. 110 sider. Faglig rapport fra Danmarks Miljøundersøgelser.
23 Dahl, K., J. Carstensen, C. Lundsgaard & J. Andersen (2001): Stenrev. 10 pp. In: Henriksen et al. (2001): Marine områder 2000. Miljøtilstand og udvikling. Faglig rapport fra Danmarks Miljøundersøgelser
22 Andersen, J., L.M. Munk & S. Pedersen (2001): Vandrammedirektivet, indhold og perspektiver (1). Vand & Jord 1/2001: 17-21. 5 pp.
21 2000 Vandmiljø-2000. Status og perspektiver for indsatsen for et renere vandmiljø. Redegørelse fra Miljøstyrelsen, nr. 7/2000. Red.: J. Andersen, T. Christensen & S. Pedersen. 48 pp.
20 Andersen, J., J. Bielecki & M. Dam (2000): Bebyggelse i det åbne land. 23 pp inkl. bilag. I: Punktkilder 1999. Orientering fra Miljøstyrelsen, nr. 16/2000.
19 Andersen, J. & J. Kirkegaard (2000): Overvågning af det danske vandmiljø, 1998-2003. Stads- & Havne-ingeniøren, august 8/2000:102-109. 5 pp.
18 NOVA-2003. Programbeskrivelse for det nationale program for overvågning af vandmiljøet 1998-2003. Redegørelse fra Miljø-styrelsen nr. 1. Red: J. Kirkegaard, T.M. Iversen & J. Andersen. 397 pp.
17 1999 Vandmiljø-99. Status for vandmiljøets tilstand i Danmark. Redegørelse fra Miljøstyrelsen, nr. 1/1999. Red.: J. Ander-sen. 128 pp.
16 1998 Evaluering af Vandmiljøplanens overvågningsprogram 1993-1997. Red: J. Andersen, T.M. Iversen & J. Kirkegaard. Notat fra Miljøstyrelsen. 1998. 48 pp.
15 1997 Andersen. J. (1998): Tilsyn med vandløb og søer. 5 pp. I: Miljøtilsyn 1997. Oversigt over kommunernes og amtskommunernes miljøtilsyn. Orientering fra Miljøstyrelsen, nr. 4/1999.
14 Andersen, J. (1997): Fjorde, kyster og åbent hav. 6 pp. I: Vandmiljø-97. Miljøtilstanden i de ferske vande samt status for det øvrige vandmiljøets tilstand i 1996. Redegørelse fra Miljøstyrelsen, nr. 4, 1997. 172 sider.
13 Andersen, J. (1997): Saltvandsbaseret fiskeopdræt. 4 pp. I: Punktkilder 1996. Vandmiljøplanens overvågningsprogram: Fag-datacenterrapport. Orientering fra Miljøstyrelsen nr. 16/1997.
12 1996 Andersen, J. (1996): Saltvandsbaseret fiskeopdræt. 3 pp. I: Punktkilder 1995. Vandmiljøplanens overvågningsprogram: Fag-datacenterrapport. Orientering fra Miljøstyrelsen nr. 16/1996.
11 Christensen, P.B., F. Møhlenberg, L.C. Lund-Hansen, J. Borum, C. Christiansen, S.E. Larsen, M.E. Hansen, J. Andersen & J. Kirkegaard (1996): Havmiljøet under forandring? Konklusioner og perspektiver fra Havforskningsprogram 90. - Havforsk-ning fra Miljøstyrelsen, nr. 61. 120 pp.
10 1995 Vandmiljø-95. Grundvandets miljøtilstand samt status for det øvrige vandmiljøs tilstand i 1994. Redegørelse fra Miljø-styrelsen, nr. 3, 1995. Red.: J. Andersen & J. Stockmarr. 156 pp.
9 Iltsvind i de danske farvande i oktober 1995. Red: J. Andersen & G. Ærtebjerg. 7 pp. 8 Iltsvind i de danske farvande i september 1995. Red: J. Andersen & G. Ærtebjerg. 7 pp. 7 Redegørelse om den miljømæssige og erhvervsmæssige betydning af saltvandsbaseret fiskeopdræt i Danmark. Redegørelse
fra Miljøstyrelsen 1995. Red.: J. Andersen & K. Hansen. 8 pp. 6 1994 Andersen, J. (1995): Havbrug og saltvandsdambrug. 4 sider. I: Punktkilder 1994. Vandmiljøplanens overvågningsprogram:
Fagdatacenterrapport. Orientering fra Miljøstyrelsen nr. 10/1995.
73
5 Vandmiljø-94. Udviklingen i belastningen fra punktkilder samt status for vandmiljøets tilstand. Redegørelse fra Miljø-styrelsen, nr. 2/1994. Red.: J. Andersen & T. Christensen. 160 pp.
4 1993 Andersen, J. (1994): Havbrug. 5 pp. I: Punktkilder 1993. Vandmiljøplanens overvågningsprogram: Fagdatacenterrapport. Orientering fra Miljøstyrelsen nr. 8/1994.
3 Nordisk havovervågningsprogram - forslag til koordinering af overvågningsaktiviteter. Ed.: J. Andersen. - Nord 1993:14. Nor-disk Ministerråd. 153 pp.
2 1991 Helmgaard, P. & J. Andersen (1991): Krebs og retningslinier for udsætning. - Vand & Miljø nr. 8/1991, side 415-417, 3 pp. 1 1990 Andersen, J. & P. Helmgaard (1990): Populationsstruktur, vækstforhold og fødebiologi hos flodkrebs Astacus astacus L..
Specialerapport. Ferskvandsbiologisk Laboratorium, Københavns Universitet. 83 pp.
Publications in other languages
2 2009 Andersen, H., J.H. Andersen, A. Erichsen, I.S. Hansen, F. Møhlenberg, & E.K. Rasmussen (2009): Danske erfarenhetar av dynamiske modeller i den marina övervakningen. Organisation, model-lösningar og ”lessons learnt”, 1998-2008. DHI rap-port til Naturvårdsverket. 75 pp. (In Swedish)
1 2007 Андерсен, Э.Х., & Д. Паулак (2007): Биогенные вещества и эвтрофикация в Балтийском море: причины, последствия, решения. Парламентская конференция Балтийского моря (ПКБМ). Russian translation of Andersen, J.H & J.T. Pawlak (2006): Nutrients and Eutrophication in the Baltic Sea. Effects / Causes / Solutions. Booklet produced for the Bal-tic Sea Parliamentary Conference. 32 pp. (In Russian)
Oral presentations (since 1st of January 2011)
8 “Økosystem-baseret forvaltning af de danske havområder: Hvordan går det? by J.H. Andersen – Danish Society for Marine Biology, 26 October 2011. Invited speaker (In Danish)
7 “Towards eutrophication target setting in 5 simple steps. A status report from the HELCOM TARGREV project’ by J.H. Ander-sen & J. Carstensen – HELCOM MONAS Committee Meeting in Vilnius, Lithuania; 5 October 2011
6 “The HELCOM TARGREV: Status and perspectives” by J.H. Andersen & J. Carstensen – “HELCOM CORESET / TARGREV Joint Advisory Board Meeting”, Warsaw (Poland); 28 June 2011.
5 “Whole system assessment – the fine art of converging indicators, quality elements and assessment principles” by J.H. Ander-sen & C. Murray – “WATERS Kick-Off Meeting”, Gothenburg (Sweden); 20 April 2011.
4 “The TARGREV project in support of the HELCOM BSAP implementation process” by J.H. Andersen, J. Carstensen & K. Dormph – “HELCOM CORESET / TARGREV Joint Advisory Board Meeting”, Berlin (Germany); 20 March 2011.
3 “Towards modelling of connectivity and invasion scenarios” by J.H. Andersen, F.T. Hansen, T. Uhrenholdt & M. Potthoff – “BWO Annual Meeting”, Newcastle (United Kingdom); 25 February 2011.
2 ”Ålegræsværktøjets forudsætninger og usikkerheder” by J.H. Andersen & J. Carstensen – “Vandplaner - usikkerheder og konsekvenser for dansk landbrug”, Nationalmuseet (Copenhagen); 8 February 2011. Invited speaker. (In Danish)
1 “Havstrategidirektivets basisanalyse – en faglig og administrativ udfordring” by J.H. Andersen – ”Havforskermøde 2011”, Fuglsøcentret (Mols); 19 January 2011. (In Danish)
March 19, 2012
74
Annex 4: Manuscripts
1. Conley, D.J., S. Markager, J.H. Andersen, T. Ellermann & L.M. Svendsen, 2002: Coastal
Eutrophication and the Danish National Aquatic Monitoring and Assessment Program. Estu-
aries 25:848-861.
2. Andersen, J.H., D.J. Conley & S. Hedal, 2004: Palaeo-ecology, reference conditions and
classification of ecological status: the EU Water Framework Directive in practice. Marine
Pollution Bulletin 49:283-290.
3. Andersen, J.H., L. Schlüter & G. Ærtebjerg, 2006: Coastal eutrophication: Recent develop-
ments in definitions and implication for monitoring strategies. Journal of Plankton Research
28(7):621-628.
4. Carstensen, J., D.J. Conley, J.H. Andersen & G. Ærtebjerg, 2006: Coastal eutrophication
and trend reversal: A Danish case study. Limnology & Oceanography 51(1-2):398-408.
5. Andersen, J.H., & D.J. Conley, 2009: Eutrophication in coastal marine ecosystems: towards
better understanding and management strategies. Hydrobiologia 621(1):1-4.
6. Andersen, J.H., C. Murray, H. Kaartokallio, P. Axe & J. Molvær, 2010: A simple method
for confidence rating of eutrophication status classifications. Marine Pollution Bulletin
60:919-924.
7. Andersen, J.H., 2010: Eutrophication. Baltic Sea Environmental Proceedings 122:16-17. In:
HELCOM, 2010: Ecosystem Health of the Baltic Sea. HELCOM Initial Holistic Assessment
2003-2007. Baltic Sea Environment Proceedings 122. 63 pp.
8. Andersen, J.H., P. Axe, H. Backer, J. Carstensen, U. Claussen, V. Fleming-Lehtinen, M.
Järvinen, H. Kaartokallio, S. Knuuttila, S. Korpinen, A. Kubiliute, M. Laamanen, E. Lysiak-
Pastuszak, G. Martin, F. Møhlenberg, C. Murray, G. Nausch, A. Norkko & A. Villnäs, 2011:
Getting the measure of eutrophication in the Baltic Sea: towards improved assessment prin-
ciples and methods. Biogeochemistry 106:137-156.
9. Korpinen, S., L. Meski, J.H. Andersen & M. Laamanen, 2012: Human pressures and their
potential impact on the Baltic Sea ecosystem. Ecological Indicators 15:105-114.
10. Laamanen, M., S. Korpinen, U.-L. Zweifel & J.H. Andersen, submitted: Ecosystem health.
Textbook chapter in “Biological Oceanography of the Baltic Sea” (Eds: P. Snoeijs, H. Schu-
bert & T. Radziejewska).
848Q 2002 Estuarine Research Federation
Estuaries Vol. 25, No. 4b, p. 848–861 August 2002
Coastal Eutrophication and the Danish National Aquatic Monitoring
and Assessment Program
DANIEL J. CONLEY1*, STIIG MARKAGER1, JESPER ANDERSEN2, THOMAS ELLERMANN3, and LARS
M. SVENDSEN4
1 Department of Marine Ecology, National Environmental Research Institute, P. O. Box 358,DK-4000 Roskilde, Denmark
2 Danish Environmental Protection Agency, Strandgade 29, DK-1401 Copenhagen, Denmark3 Department of Atmospheric Environment, National Environmental Research Institute, P. O. Box
358, DK-4000 Roskilde, Denmark4 Environmental Monitoring Coordination Section, National Environmental Research Institute,
P. O. Box 358, DK-4000 Roskilde, Denmark
ABSTRACT: Nutrient over-enrichment and cultural eutrophication are significant problems in the Danish marine en-vironment. Symptoms of eutrophication include periods of hypoxia and anoxia in bottom waters, death of benthic-dwelling organisms during anoxia, long-term reductions in the depth distribution of macrophyte communities, changesin the species composition of macrophyte communities, and increases in reports of harmful algal blooms. In 1987 theAction Plan on the Aquatic Environment was adopted to combat nutrient pollution of the aquatic environment with theoverall goal of reducing nitrogen loads by 50% and point source phosphorus loads by 80%. The Danish Aquatic Nation-wide Monitoring Program was begun in 1988 in order to describe the status of point sources (industry, sewage treatmentplants, stormwater outfalls, scattered dwellings, and fish farms), ground water, springs, agricultural watersheds, streams,lakes, atmospheric deposition, and the marine environment. Another important aspect of the program was to documentthe effects on the aquatic environment of the measures and investments taken for nutrient reduction as outlined in theAction Plan. The monitoring program should determine if reductions in nutrients are achieved by the measures takenand should help decision makers choose appropriate additional measures to fulfill the objectives. Coordination withinternational programs and commissions is an important component of the monitoring program to meet internationallyagreed upon reductions in nutrient inputs. The future and direction of the Danish National Aquatic Monitoring andAssessment Program will be to a large extent shaped by both the Water Framework Directive and Habitat Directiveadopted by the European Union.
IntroductionNutrient over-enrichment and cultural eutrophi-
cation of surface waters have been recognized asbeing one of the most urgent problems to over-come in order to improve the environmental stateof lakes, streams, and marine waters in Denmark.Nutrient loading from Denmark expressed on anarea basis ranks among the highest in Europe (Paa-by and Møhlenberg 1996) and reflects the densityof the population and the intensity of agriculture.Danish marine waters display all the classic symp-toms associated with eutrophication including in-creased phytoplankton biomass and harmful algalblooms (Kaas et al. 1999), reductions in the depthdistribution of macrophyte communities (Sand-Jensen et al. 1994), changes in the species com-position of macrophyte communities (Middleboeand Sand-Jensen 2000), periods of hypoxia and an-oxia in bottom waters (Conley and Josefson 2001),
* Corresponding author: tele: 145 4630 1200; fax: 145 46301114; e-mail: [email protected].
and the death of benthic-dwelling organisms dur-ing anoxia (Fallesen et al. 2000).
During the 1980s a series of oxygen (O2) deple-tion events catalyzed the environmental movementand led to the passage of the Action Plan on theAquatic Environment (Christensen et al. 1998). Inaddition to creating a national agenda to reducethe over-enrichment of natural waters with nutri-ents, the Action Plan established a coordinatedAquatic Nationwide Monitoring Program (Kron-vang et al. 1993) leading to the creation of theNational Environmental Research Institute(NERI). The 1987 Action Plan on the Aquatic En-vironment I was only the first of a series of legis-lative actions in combination with internationalagreements that have arisen to reduce nutrientover-enrichment and the symptoms of eutrophi-cation in Denmark with the overall objective of aclean and healthy aquatic environment. Most ofthe Action Plans are concerned with reduction ofdischarges, loads, and emission of nutrients. Thecurrent Danish National Aquatic Monitoring and
Danish Monitoring and Assessment Program 849
Assessment Program which began in 1998 is de-signed to run through 2003 and is probably oneof the most comprehensive programs globally.
The purpose of this paper is to provide an over-view of the current efforts in Denmark to reducenutrient over-enrichment of marine waters and tomonitor the aquatic environment. We present thevarious National Action Plans and relevant parts ofderived legislation from the last 15 years that aimto reduce nutrient inputs, we provide an overviewof the Danish National Aquatic Monitoring and As-sessment Program focusing on the marine environ-ment and the measurement of nutrient inputs, andwe briefly examine the documented changes in nu-trient loading and the effects on the aquatic envi-ronment. We also address various aspects of themonitoring program.
Strategies and Measures to Reduce NutrientInputs to the Danish Aquatic Environment
POLITICAL OBJECTIVES
Since the mid-1980s a high priority has been giv-en to the quality and protection of ground waterand surface water in Denmark with an overall goalof ensuring that the waters are clean. The endeav-ors are described in the Environmental PolicyWhite Paper (Ministry of Environment and Energy1999), which states that the Government will worktowards ensuring that streams, lakes, and marinewaters are clean and of a satisfactory quality withregard to health and hygiene; that exploitation ofthe water bodies and associated resources takesplace in a sustainable manner; and that the objec-tives of relevant international agreements will befulfilled. The central legal instrument to fulfillthese political objectives is the Consolidated Envi-ronmental Protection Act, which aims to safeguardthe environment, to support a sustainable socialdevelopment, and to conserve the flora and fauna(Ministry of Environment and Energy 1998).
Overall objectives for streams state that waterflow must be adequate, that obstructions must nothinder the dispersal of fish and macroinverte-brates, that there shall be 2-m wide borders free ofcultivation along natural streams, that streams havegood oxygen conditions, and that they contain avaried and natural fauna and flora. Overall objec-tives for Danish lakes are that animal and plantcommunities should be natural, that the watershould be clear and that submerged macrophytesshould be present in the shallow parts of the lakes(Ministry of Environment and Energy 1999).
The overall objectives for Danish marine watersare based on the 1992 Helsinki Convention on Pro-tection of the Marine Environment in the BalticSea Region (HELCOM 1992), the 1992 Conven-
tion for the Protection of the Marine Environmentof the Northeast Atlantic (OSPAR 1992), and the1995 Declaration of the 4th International Confer-ence on the Protection of the North Sea (DanishEnvironmental Protection Agency 1995). The over-all objectives are that the fauna and flora may onlybe insignificantly or slightly affected by anthropo-genic pollution and human activities, that nutrientlevels have to be at a natural level, the clarity ofthe water has to be normal, unnatural blooms oftoxic phytoplankton or pollution-dependent ma-croalgae must not occur, and oxygen deficiencymay only occur in areas where it is natural, andthat the levels of hazardous substances have to beat background levels in the case of naturally oc-curring substances and close to zero in the case ofhazardous substances. Commercial exploitation(fisheries, navigation, drainage, offshore industry,minerals extraction, marine dumping of seabedmaterial, recreational activities, and other uses ofsurface water) has to be conducted in a mannerthat respects environmental and natural wealthand is sustainable (Ministry of Environment andEnergy 1999).
STRATEGIES AND MEASURES
The primary means of achieving the quality ob-jectives for both ground water and surface watersare reductions in nutrient loads. On November 28,1986 the Danish Parliament adopted an agendathat instructed the Government to reduce totalloads of nitrogen (N) and phosphorus (P) to theaquatic environment by 50% and 80%, respective-ly. These reductions correspond to a change in an-nual loads from a level of 283,000 tonnes N and9,120 tonnes P at the time the plan was adoptedto levels of about 141,600 tonnes N and 1,820tonnes P (Folketinget 1987). During the politicalprocess in 1987, these overall reduction targetschanged from being targets for loads to the aquaticenvironment to reduction targets for the discharg-es and losses from three sectors: agriculture, mu-nicipal wastewater treatment plants (WWTP), andpoint industrial discharges. It could be argued thatthis change was rational since these three sectorsare the most relevant to reducing eutrophication.As a result of the change, which was not in accor-dance with the November 28 Agenda, the adoptedAction Plan on the Aquatic Environment did notinclude a reduction target for emissions of N tothe atmosphere, reduction targets for dischargesor losses from aquaculture, scattered settlements,stormwater overflows, and offshore activities, anda reduction target for P losses from agriculturalfields.
It is widely recognized that pollution of marinewaters crosses political boundaries. The countries
850 D. Conley et al.
around Denmark in the North Sea and Baltic Seaadopted similar reduction targets through threedifferent conventions. The North Sea Conferencein London in 1987 (the countries of the North Searegion excluding the United Kingdom) adoptedthe goal of reducing N and P inputs to the sea by50% over the period 1985–1995 in areas wherethese could cause pollution. These reduction tar-gets were reiterated at the conferences in TheHague in 1990 and Esbjerg in 1995. The ParisCommission in June 1988 adopted a 50% reduc-tion target for nutrient inputs to marine waters sus-ceptible to eutrophication and also adopted a pro-gram to achieve the reductions. In 1989, the re-duction targets were specified in relation to specif-ic sectors. In 1992, it was decided to integrate theOslo and Paris Conventions (the OSPAR Conven-tion), both of which aimed to prevent marine pol-lution of the Northeast Atlantic region from dump-ing and land-based sources of pollution. In Feb-ruary 1988 the Helsinki Commission (HELCOM)adopted a declaration specifying a 50% reductiontarget for discharges of polluting substances, in-cluding nutrients, over a 10-yr period. In 1998, theministers confirmed their commitment to attain-ing the strategic goal from 1988 and defined spe-cific objectives that must be achieved before theyear 2005.
Enacted Measures Taken to Achieve NutrientReductions
SPECIFIC REDUCTION TARGETS FOR THE
AGRICULTURAL SECTOR
Since the mid-1980s, a number of action plansand strategies have been adopted by the DanishParliament to regulate development of the agri-cultural sector, one of the main sources of nutri-ents to the aquatic environment. The action plansinclude the NPo (nitrogen, phosphorus, and or-ganic matter) Action Plan in 1985, the Action Planon the Aquatic Environment I in 1987, the ActionPlan for Sustainable Agriculture in 1991, parts ofthe Government’s 10-Point Program for Protectionof the Ground Water and Drinking Water in 1994,follow-up on the Action Plan for Sustainable Agri-culture in 1996, the Action Plan on the AquaticEnvironment II in 1998, and the Agreement onMay 2, 2001 on Supplementing Initiatives andPreparations for the Action Plan III.
The reduction targets for N and P stipulated inthe Action Plan on the Aquatic Environment I arean approximate halving of point source N loadsand a 80% reduction of point source P loads, in-cluding the elimination of the P farmyard load.The reduction targets were to be attained by 1993through the following measures carried out by the
agricultural sector: establishment of sufficient ca-pacity to store 9 mo of manure production so thatmanure can be stored until the crop growth seasonbegins, establish crop rotation and fertilizationplans to ensure that the N content of fertilizer isoptimally exploited, agricultural fields must havegreen cover during the winter period, manure hasto be plowed in or in some other way deployedinto the soil within 12 h of application, and limitson the amount of livestock manure applied to ag-ricultural fields (Table 1).
It soon became clear that it would not be pos-sible to attain the reduction targets by 1993 (Min-istry of Agriculture 1991). The measures stipulatedin the Action Plan on the Aquatic Environment Iwere tightened in 1991 in the Action Plan for Sus-tainable Agriculture. The reduction target wasmaintained, but the time frame was extended tothe year 2000. The measures were fertilization ac-counts so that fertilizer application could be doc-umented; more stringent and fixed requirementson utilization of the N content of livestock manure;all farms must establish sufficient capacity to store9 mo of manure production; and a ban on theapplication of liquid manure between harvest timeand February except on agricultural fields cultivat-ed with winter rape or grass. Since the Action Planfor Sustainable Agriculture there have been a num-ber of follow-up plans for reducing the impact ofthe agricultural sector, including the Govern-ment’s 1994 10-Point Program for Protection ofthe Ground Water and Drinking Water in Den-mark.
The need to further tighten the regulation ofagricultural loads of N has become even more nec-essary because Denmark must comply with the Eu-ropean Union (EU) Nitrates Directive by the year2003. The directive restricts the application of live-stock manure to 170 kg N ha21 yr21. In the case ofsome farms this is less than the levels currently per-mitted. Denmark has sought permission to deviatefrom the 170 kg N ha21 yr21 rule on cattle holdingsto enable the application of up to 230 kg N ha21
yr21 on a small number of these holdings.In February 1998, the Danish Parliament adopt-
ed several new instruments aimed at achieving thereduction targets. The Action Plan on the AquaticEnvironment II will reduce N leaching by a further37,000 tonnes N yr21 to enable the reduction targetof 100,000 tonnes N yr21 to be achieved no laterthan the end of the year 2003 (Table 1; DanishEnvironmental Protection Agency 2000). Underthe Action Plan on the Aquatic Environment II,16,000 ha of wet meadow will be re-established tohelp reduce N leaching through denitrification,20,000 ha forest will be planted before 2002, andagri-environmental measures that include financial
Danish Monitoring and Assessment Program 851
TABLE 1. Summary of measures and estimated reductions (tonnes N yr21) in nitrogen loading from agriculture. The years 1993,2000, and 2003 are the expected year of implementation of Action Plan on the Aquatic Environment I; Action Plan for SustainableAgricultural Development; and Action on the Aquatic Environment II, respectively. The reduction figures are not legally binding,they should for that reason be considered merely as politically agreed upon goals. The goal of Action Plan II is 100,000 tonnes Nyr21; therefore the farmyard load (27,000 tonnes N yr21) is added to the estimated effect of Action Plan I and Action Plan forSustainable Agricultural Development (89,900 2 27,000 5 62,900 tonnes N yr21).
Action Plans and Measures: 1993 2000 2003
Action Plan on the Aquatic Environment I (1987):Optimal utilization of livestock manureNPo Action PlanNPo Subsidy ActFurther initiatives
55,0005,000
10,000Program for improved utilization of fertilizersSystematic fertilization plansImproved application methodsWinter green fields—catch crops and plowing down of straw
15,0005,000
20,000Winter green fields—further initiativesStructural measuresTotal
Action Plan for Sustainable Agricultural Development (1991):
8,0009,000
127,000 50,000
Improved utilization of livestock manureReduction in commercial fertilizer consumptionProtection of groundwater in particularly vulnerable areasReduction in agricultural acreageStructural development, other measuresTotal
20,000–40,0008,000–15,0001,000–2,000
17,000–20,00015,00077,000 89,900
Action Plan on the Aquatic Environment II (1998):WetlandsSensitive agricultural areasAfforestationImproved fodder utilizationStricter harmonization criteria
5,6001,9001,1002,400
300Stricter requirements on utilization of N content of manureOrganic farmingCatch crops on a further 6% of the land10% reduction in N normTotal
Total: 127,000 127,000
10,6001,7003,000
10,50037,100
127,000
support to farmers willing to cultivate sensitive ag-ricultural areas in a more environmentally soundmanner will be implemented. Agricultural mea-sures include using less fertilizer or completely re-fraining from cultivating the land (there has hith-erto been very little interest in this scheme), im-proved fodder utilization and changes in feedingpractice, implementation of stricter harmony cri-teria governing livestock density, stricter require-ments on utilization of the N content of livestockmanure, converting 170,000 ha to organic farming,catch crops on a further 6% of a farmer’s land,and reducing the N norm by 10%, e.g., farmersmay now only apply N in amounts correspondingto 90% of the economically optimal level.
If the measures in the Action Plan on the Aquat-ic Environment II are fully implemented, it is ex-pected that within several years it will result in a100,000 tonnes N yr21 reduction in leaching fromagricultural land. N consumption in the form ofcommercial fertilizer will decrease from 400,000tonnes N yr21 in 1985 to 200,000 tonnes N yr21 in2003 (Iversen et al. 1998).
In connection with the Action Plan on theAquatic Environment I, it was estimated that Nloads could be reduced by a total of 127,000 tonnesN yr21 by 1993. The reduction targets were 100,000tonnes N yr21 for the N load from agriculturalfields and 27,000 tonnes N yr21 for the farmyardload. In the Action Plan for Sustainable Agricul-ture it was estimated that by the year 2000 the mea-sures stipulated in the Action Plan on the AquaticEnvironment I would only have reduced N loadsby 50,000 tonnes N yr21 and that further measureswere needed. The existing measures and targetswere re-evaluated in 1998 in connection with thepreparation of the Action Plan on the Aquatic En-vironment II, and it was concluded that by the year2003 the existing measures would reduce N loadsby 89,900 tonnes N yr21. Together with the ex-pected reduction under the Action Plan on theAquatic Environment II, it was concluded that Nloads would be reduced by 127,000 tonnes N yr21
by 2003.Not all the measures in the Action Plan on the
Aquatic Environment II will have taken full effect
852 D. Conley et al.
by 2003. A mid-term evaluation in late 2000 indi-cated that further measures were needed to fulfillthe 100,000 tonnes N yr21 reduction of root zonelosses. It is assumed that an Action Plan III will bepassed in the Danish Parliament in 2003 or 2004with regional reduction targets including measuresagainst diffuse P losses.
SPECIFIC REDUCTION TARGETS FOR MUNICIPALWASTEWATER TREATMENT PLANTS
Discharges from municipal WWTP are regulatedby the Environmental Protection Act and deriva-tive statutory orders. The EU Council Directive91/271/EEC concerning Urban Wastewater Treat-ment as amended by Commission Directive 98/15/EU, commonly referred to as the Urban Waste-water Directive, is one of the most important legaldocuments in the EU concerning water quality.The purpose of the directive is to protect the en-vironment against the negative effects associatedwith the discharge of inadequately treated urbanwastewater and discharges of biologically degrad-able industrial wastewater from the food process-ing industry. According to the directive, wastewaterdischarges must be subjected to a level of treat-ment appropriate to the environment in questionand to the designated use of the recipient waterbody. Denmark implemented the provisions of thedirective in 1994 legislation.
The Action Plan on the Aquatic Environment’sreduction targets for municipal WWTP were ad-justed in 1990 on the basis of the results of theNationwide Aquatic Monitoring Program (DanishEnvironmental Protection Agency 1991). In thecase of N, annual discharges in treated wastewaterwere to be reduced from 18,000 tonnes N yr21 to6,600 tonnes N yr21. P discharges were reducedfrom 4,470 tonnes P yr21 to 1,220 tonnes P yr21.The reduction in N corresponds to all new or up-graded plants exceeding 5,000 personal equiva-lents (PE) and all existing plants exceeding 1,000PE having to implement biological treatment withN removal down to an annual average of 8 mg Nl21. Municipal WWTP exceeding 5,000 PE have toremove P down to an annual average of 1.5 mg Pl21.
SPECIFIC REDUCTION TARGETS FOR POINTINDUSTRIAL DISCHARGES
Point discharges from industry are regulated bythe Consolidated Environmental Protection Actand the EU Directive on Pollution Prevention andControl (IPPC Directive) and derivative statutoryorders. The IPPC Directive aims at integrated pre-vention and control of pollution by major indus-tries. The directive specifically regulates the energyindustry (e.g., power stations and refineries), pro-
duction and processing of metals, the mineral in-dustry, the chemical industry, waste managementplus a number of other activities such as papermanufacturers, textiles pre-treatment and dyeing,slaughterhouses and dairies, and installations forintensive rearing of poultry and pigs exceeding acertain capacity. The IPPC Directive contains mea-sures designed to prevent or, where that is notpracticable, to reduce emissions to the air, water,and land. Because of the large differences betweenindividual enterprises and their discharges ofwastewater, the Action Plan on the Aquatic Envi-ronment I did not stipulate general discharge re-quirements for industry as it did for WWTP. In-dustry was to reduce its discharges through the ap-plication of Best Available Technology (BAT) atthe level of treatment that is technically attainableand economically viable.
Costs of Action PlansThe costs of Action Plan on the Aquatic Envi-
ronment I for the period 1985–1989 have been es-timated in 1990 to be 1.2 billion C5 (Danish Envi-ronmental Protection Agency 1991). This figure in-cludes some investments agreed upon already inthe 1985 NPo Action Plan and reaffirmed in the1987 Action Plan. The investments in agricultureduring the period 1985–1992 have been 400 mil-lion C5 with total investments in municipal WWTPof 1.1 billion C5 including both construction of newplants and enlargement and improvement of ex-isting plants. In addition, industries with separatetreatment and discharge of wastewater invested135 million C5 in improved wastewater treatment.The cost of Action Plan II is expected to be 135million C5 , half of which will be financed by theagricultural sector (Iversen et al. 1998). The ex-penses of wetland restoration, groundwater protec-tion areas, afforestation, and development of or-ganic farming will be financed by different govern-ment agencies.
The total annual costs are 1.28 billion C5 total forplanning and management of the aquatic environ-ment by state, regional, and local authorities: 68billion C5 , maintenance and restoration of rivers,streams and lakes: 81 million C5 , national monitor-ing of the aquatic environment: 40 million C5 , su-pervision by the counties and local communities:11 million C5 , supply of clean and healthy drinkingwater: 405 million C5 , and discharge and cleaningof wastewater from households and industries: 676million C5 (Christensen personal communication).
The Danish National Aquatic Monitoring andAssessment Program
MONITORING OF ATMOSPHERIC DEPOSITION OFNUTRIENTS TO DANISH MARINE WATERS
The atmospheric component of the Action Planon the Aquatic Environment was initiated in 1989
Danish Monitoring and Assessment Program 853
Fig. 1. The geographical distribution of sampling stations invarious parts of the Danish Aquatic Monitoring and AssessmentProgram. A) Atmospheric measurement stations. B) Catchmentdistribution used to measure watershed input of nutrients. C)Marine monitoring stations.
with the focus of determining atmospheric N de-position to Danish waters. The first monitoringprogram (1989–1994) expanded upon existingmonitoring stations for better geographical cover-age. The program consisted of 6 stations for wetand dry deposition of different N species and 12stations where only wet deposition was measured.About half of the stations were placed close to thecoast. Interpolations of measurements between sta-tions were used to obtain the overall deposition toDanish waters.
The program consists of sampling at the mainstations for wet deposition with bulk samplers ona half-month basis. The precipitation samples areanalyzed for their content of nutrients (mainly am-monium and nitrate) and a number of other airpollutants (including 9 heavy metals). The gasphase and particulate phase air pollutants are col-lected on a daily basis on various types of filters,which after extraction and analysis are used to de-termine the atmospheric content of N (ammonia,particulate ammonium, and the sum of nitric acidand particulate nitrate) and other important airpollutants. Dry deposition at the monitoring sta-tions is subsequently calculated by use of literaturedata for the deposition velocities and measuredconcentrations and meteorology.
During the early 1990s a comprehensive modelfor calculation of deposition to Danish marine wa-ters was developed, the ACDEP-model (Atmo-spheric Chemistry and Deposition; Hertel et al.1995). The ACDEP model is a trajectory modelthat calculates the atmospheric concentrations andthe wet and dry deposition of nutrients and otherimportant air pollutants to 30 3 30 km grids cov-ering Danish marine waters and land. The trans-port of air pollutants is determined by using 96-hback trajectories calculated by use of meteorolog-ical data from NERI (Brandt et al. 2000) and Co-operative Programme for Monitoring and Evalua-tion of the Long Range Transmission of Air Pol-lutants in Europe (EMEP). The model is suppliedwith information of initial concentrations based ona coarser long-range transport model (Brandt etal. 2000) and emissions (from NERI and EMEP)of the air pollutants, and simulates the vertical andhorizontal transport, chemical transformations (80reactions), and the wet and dry deposition of 37air pollutants.
In 1995 model calculations were implemented asan integral part of the monitoring program and atthe same time the number of precipitation stationswere reduced to only two stations (Fig. 1a). Theaim was to improve the results through a combi-nation of both measurements and model calcula-tions. The measurements at the monitoring sta-tions are used to determine the concentrations
854 D. Conley et al.
Fig. 2. Long-term annual means for the concentration ofammonia (closed squares), particulate ammonium (circles),and sum of nitric acid and particulate nitrate (sum nitrate; opensquares) in the atmosphere at the monitoring station at Anholt.
Fig. 3. ACDEP Model calculation of the annual total deposition of nitrogen (tonnes N km22) to Danish marine waters in 1999.Note that the numbers are only valid for the part of the grid with water surface (Ellermann et al. 2000).
and deposition at monitoring stations, while themodel calculations are used to calculate the de-position to both land surfaces and Danish marinewaters. Subsequently, measurements are used tovalidate the results from model calculations. Sea-sonal and long-term trends are determined solelyon the basis of measurements due to lack of suf-ficiently updated emission data, while studies of
the origin of air pollution are based on model cal-culations. The influence of Danish emissions canbe estimated by model calculations with and with-out Danish emissions.
Long-term trends in concentrations at Anholt,situated in the middle of Kattegat (Fig. 1a), showa significant decrease of particulate ammoniumand sum nitrate, whereas long-term trends in am-monia concentration are insignificant (Fig. 2).Similar patterns are observed at other Danishmonitoring stations (Ellermann et al. 2000). Re-ductions in emissions of N compounds have oc-curred in central Europe and Denmark from re-ductions in both combustion processes and farm-ing. Calculation of the annual total deposition ofN to Danish marine waters in 1999 from the AC-DEP model shows that N deposition varies between1.0 and 1.8 tonnes N km22 with an average of 1.1tonnes N km22 for all the Danish waters (Fig. 3).Moreover, model calculations show that the highN deposition rates to estuaries and coastal watersare mainly due to the high dry deposition of am-monia from local animal husbandry.
MONITORING OF LAND-BASED NUTRIENT LOADS TODANISH MARINE WATERS
Denmark uses a load-oriented approach to quan-tify the load of nutrients from land to coastal areas.The method is based on an extensive monitoringprogram in rivers and streams and on all pointsources larger than 30 PE. WWTP, single industriesnot connected to WWTP, aquaculture, and somestormwater outfalls are also monitored. Nutrientloads from scattered dwellings in rural areas, the
Danish Monitoring and Assessment Program 855
remaining stormwater outfalls, and other pointsources less than 30 PE are estimated by standardvalues or models. A characteristic feature of theDanish landscape is the lack of large rivers. Be-cause it is economically and practically impossibleto monitor the several thousand small streams thatcover the landscape, the country has been dividedinto monitored and unmonitored regions in rela-tion to deriving riverine loads.
Nutrient loads are determined using approxi-mately 130 river monitoring stations situated down-stream in rivers. About 55% of the land surfacearea is covered by riverine measurements. Watersamples are taken generally between 12 and 26 (av-erage 19) times per year, the stage is continuouslyrecorded, and stage-discharge relationships areused to calculate load. In addition to monitoringstations in rivers located near the coast, 150 sam-pling stations are situated in small agriculturalcatchments (5 to 60 km2) with only minor inputsfrom point sources. The loads in the unmonitoredpart of the country are calculated using flow-weighted concentrations and discharge from agri-cultural catchments with corresponding climate,soil type, geology, and agricultural practices asfound in the monitored catchments, and then add-ing monitored point sources.
The total loads to coastal areas are determinedby summing the total monitored load, the total un-monitored load, and nutrient loads from pointsources discharging directly to coastal areas (Fig.1b). Source apportionment is performed on thetotal load to evaluate the importance of differentnutrient sources. Nutrient loads from diffusesources, such as agricultural land and forested ar-eas, are estimated as the difference between thegross transport, including retention in rivers andlakes, and the total load from point sources. Nu-trient loads from cultivated areas include the esti-mated potential load from scattered dwellings en-tering the surface freshwater system.
Natural background loads of N and P constitutea part of the total estimated nutrient inputs to sur-face water and include loads from unmanagedland and that part of the loads of N and P frommanaged land that would occur irrespective of an-thropogenic activities. The natural backgroundloads are determined from measurements of nu-trient loads in 9 small non-agricultural catchmentsand subdivided into sandy and loamy catchments.The means for these 9 natural watersheds from1989 to 1999 were 1.50 6 0.15 mg N l21 (2.20 60.76 kg N ha21) and 50 6 6 Tg P l21 (0.078 6 0.04kg P ha21).
Nutrient loads to freshwater are often greaterthan the measured nutrient transport to coastal wa-ters due to the retention of nutrients in lakes and
rivers. Retention plays a key role for the amountand the composition of nutrient fluxes, especiallyP, through river systems (Billen et al. 1991). Reten-tion in the catchment must be added to the mea-sured load to estimate the amount of diffusesource (natural background 1 agriculture 1 scat-tered dwellings) loads to freshwater. In larger riv-ers, N loss through denitrification can have a sig-nificant influence on the total load from the riversystem. In-stream river retention of N in Danishstreams can be high, primarily during the summer,but over an annual cycle is negligible and there-fore not included in the calculations. The reten-tion of P is especially important in streams whereover bank flooding occurs. For the total load fromDenmark to coastal areas the net retention of Nand P in rivers and streams is less than 1% to 2%(Svendsen et al. 1995). Retention in lakes consti-tutes 8% to 12% of the gross riverine N load and0% to 2% of the gross riverine P load from 1989–1999 (Svendsen et al. 1998). The retention in lakesis calculated from mass balance calculations inabout 30 intensively monitored lakes.
The measures taken to reduce the load frompoint sources in Denmark have been successful.Since 1989 the N load from point sources has beenreduced by 66% and for P by 81% (Fig. 4). Themain part of the reduction has taken place by im-proved and extended purification of wastewaterfrom WWTP and from single industrial plants notconnected to treatment plants (SID). The N loadfrom WWTP has been reduced by 74% and for Pby 90% and the corresponding figures for SID are85% and 95%, respectively. Further, the use of de-tergents without P reduced the P load from scat-tered dwellings. Thus, the original reduction tar-gets for P have been fulfilled.
The total load of N to coastal waters is closelyrelated to runoff (Fig. 5). P loads have been dom-inated by inputs via wastewater load, but after 1994the diffuse loading of P has been a significant por-tion of the load to coastal areas. The reduction inthe contribution of point sources means that Ploading is now correlated to runoff. Diffuse sourc-es are the biggest N source. In 1999 81% of thetotal N load originated from agricultural sources,11% from natural background loads, and 8% frompoint source loads. There has been a significantreduction (Kendall trend test, p , 0.01) in the to-tal P load to coastal areas. There is a tendency ofa minor reduction in diffuse N loads, but it is notsignificant (p . 0.05).
Marine MonitoringA marine component of the Aquatic Nationwide
Monitoring Program was initiated in October 1988with implementation of the Action Plan for the
856 D. Conley et al.
Fig. 4. Freshwater discharge, total nitrogen load, and totalphosphorus load to coastal waters in Denmark divided up bysource. WWTP are wastewater treatment plants, SID are singleindustries not connected to treatment plants, SWO are storm-water overflows, SD are scattered dwellings, and AQ are fresh-water and coastal fish farms.
Fig. 5. Freshwater discharge, total nitrogen load, and totalphosphorus load to coastal waters in Denmark from differentsources. Diffuse loads also include natural background loadsand loads from scattered dwellings, freshwater point sources arethose discharging directly to freshwaters and direct point sourc-es are those discharging directly into estuaries or coastal waters.
Aquatic Environment I. The program was based onprevious national and regional monitoring activi-ties. Involvement of regional authorities in themonitoring activities broadened the expertiseavailable to local decision makers and reducedship time for high frequency sampling. All data arestored in a national database at the Marine TopicCentre, National Environmental Research Insti-tute, Roskilde, Denmark, whose staff is responsiblefor writing technical guidelines, coordinating themarine monitoring program, and producing anannual national report about the State of the Ma-rine Environment.
Marine monitoring consists of several compo-nents. Water column physical, chemical, and bio-logical parameters are measured 26 to 46 times peryear at 16 open water and 38 estuarine stations.Another 26 open water stations are monitored fordissolved oxygen (DO) four times in the fall, andtwice a year at all 92 stations including the Danish
portion of the North Sea and Skagerrak (Fig. 1c).Physical and chemical parameters include conduc-tivity, temperature, depth, light, nutrients, and ox-ygen. Oxygen concentration data provide infor-mation about the distribution of oxygen depletionas a consequence of eutrophication (Conley andJosefson 2001). Species composition, coverage anddepth distribution of angiosperms and macroalgaeare assessed by scuba diving in 34 estuaries and 9stone reefs in open waters (Middleboe et al. 1998).Every third year the distribution in shallow watersis assessed with the use of aerial photography. Thebiomass and species composition of benthic faunais monitored at 24 stations in estuaries and 24 sta-tions in open waters once a year ( Josefson andRassmussen 2000). Biomass, distribution, and con-dition of benthic filter feeders are monitored infive estuaries. Measurements of sediment chemis-try include pools of nutrients, H2S concentrationsat different depths, fluxes of nutrients between the
Danish Monitoring and Assessment Program 857
TABLE 2. Cost of the Danish Aquatic Monitoring and Assessment Program 1998–2003. Prices in million € per year (1996 prices).
Counties State Institutions Total
Agricultural catchmentsGround waterSprings, watercourses, lakesPoint sourcesAtmospheric deposition
1.953.625.224.270
0.200.230.600.301.28
2.153.855.824.571.28
Marine watersCrosscutting research and developmentCoordination and administrationTotal
5.92——
21.0
1.300.700.314.92
7.220.700.31
25.9
sediment and the overlying water, and measure-ments of denitrification. Hydrographic modeling isconducted for the entire area with a three-dimen-sional model with the aim of calculating nutrientfluxes between the Baltic Sea, the North Sea, andthe Kattegat. Separate models for water exchangeare run on six type area estuaries.
The marine program is divided into two parts:estuaries and open marine areas. Six estuariesserve as type areas for the categories of Danish es-tuaries (Horsens Fjord, Limfjorden, Odense Fjord,Ringkøbing Fjord, Roskilde Fjord, Skive Fjord).They are monitored intensively, up to 46 times peryear, for most water column parameters mentionedabove. A broader geographic coverage includes 34additional estuaries but at a lower frequency ofsampling (Fig. 1c). The open marine area consistsof 16 intensively sampled water column stations, 76additional stations at a lower frequency, a three-dimensional hydrographic model, and monitoringof macroalgae on 9 stone reefs. Data from themore frequently sampled stations support a de-tailed understanding of the mechanisms relevantto eutrophication in response to nutrient loadchanges within a context of climatic variation. Datafrom the broader geographic coverage are used todocument long-term changes in winter concentra-tions of nutrients and the fall distribution of oxy-gen depletion.
Costs of Monitoring ProgramThe total annual costs of the Danish National
Aquatic Monitoring and Assessment Program areabout 26 million C5 (see Table 2), apportioned be-tween national agencies (5 million C5 yr21) and re-gional authorities (21 million C5 yr21). In parallelto the activities of the Danish National AquaticMonitoring and Assessment Program, the Danishcounties have a number of compliance monitoringactivities in order to document that legally bindingdischarge permits are kept and assess the devel-opment and fulfillment of normative ecologicalquality objectives. The annual costs of compliancemonitoring are about 11 million C5 yr21 (Andersenet al. 2001).
Effects of the Action Plan on Marine Waters
The activities of the Aquatic Nationwide Moni-toring Program (1988–1997) and the follow-upprogram, the Danish National Aquatic Monitoringand Assessment Program (1998–2003) have beeninstrumental in documenting eutrophication ef-fects in Danish waters (e.g., Ophelia 1995a,b;Jørgensen and Richardson 1996; Christensen et al.1998; Conley et al. 2000). New results regardinglong-term trends in the marine environment fromthe Danish National Aquatic Monitoring and As-sessment Program are presented here.
Nutrient management actions have resulted in asignificant decrease in the P load from point sourc-es and a decrease (not significant) in the N load(Fig. 5). The reductions come as a result of gradualincreases in the tertiary treatment of sewage firstwith implementation of local plans in the 1980s toreduce the effects of eutrophication in inland wa-ters, and later with implementation of the 1987 Ac-tion Plan on the Aquatic Environment. Conse-quently, the potential for nutrient limitation hasincreased, especially in the spring for P and signif-icant reductions have occurred in nuisance growthof macrophytes, e.g., Ulva lactua (Conley et al.2000).
Significant relationships were found betweenloads of N and P and the mean Secchi disk depthfrom 1989 to 1999 (Fig. 6). There is a significantlinear correlation between both N loading and Ploading in estuaries with Secchi disk depth (p 50.006 for P and p 5 0.026 for N). In the open areasonly N is significantly correlated to Secchi diskdepth (p 5 0.14 for P and p 5 0.001 for N). Therelationship for P appears somewhat curvilinearperhaps indicating that the Danish estuaries weresaturated with P at loadings above 4,000 tonnesyr21. The data suggest that P is limiting for primaryproduction in estuaries, at least during part of theseason, and extensive P limitation is supported byexperimental data from Danish estuaries (Holm-boe et al. 1999). Secchi depth increased 30% froma level of 3 m between 1984 and 1988 to 3.9 mbetween 1996 to 1999, while over the same time
858 D. Conley et al.
Fig. 6. The relationship between Secchi disk depth and the total land-based loads of phosphorus and nitrogen from Denmarkbetween 1989 and 1999 in estuaries and open sea areas. Estuaries also include near coastal areas with a water depth less than 10 m.Open waters are all areas outside the estuaries with a water depth greater than 10 m.
period areal primary production decreased by 28%(Fig. 7).
While implementation of nutrient reductionsmandated in the Action Plans are legally enforce-able, the effects of the Action Plans on nutrientloading are not expected to occur within the timeframe of the implementation of the measures tak-en. This is due to the fact that there are N and Pstores in soils, ground waters, and marine sedi-ments that must be depleted before actual nutrientreductions may be observed in the environment.Reductions in nutrient loads are expected to even-tually occur when all measures are implemented,but it is doubtful whether the target of a 50% re-duction in N load will be achieved. Positive effectsin the marine environment are currently docu-mented, and further improvements are expected.A practical example of the effects of nutrient re-ductions occurred during the two dry years of 1996and 1997 with a 50% reduction in N loading to themarine environment. Effects on the marine envi-ronment included lower than average water col-umn chlorophyll concentrations, increased depthpenetration of aquatic macrophytes, and reducedextent of oxygen depletion (Markager et al. 1999;Rask et al. 1999).
General Aspects of the Monitoring ProgramThe Danish National Aquatic Monitoring and
Assessment Program illustrates the importance ofmonitoring data feedback upon regulatory actionsfor the successful management of the aquatic en-vironment. As shown earlier, evaluation of the Ac-tion Plan on the Aquatic Environment I demon-strated that the measures taken were not sufficientto reduce nutrient levels to the targeted amounts.Additional measures to reduce nutrient loads weretaken with the Action Plan on the Aquatic Envi-ronment II. The process is also an example of stra-tegic environmental planning (Table 1), wherebya target is set, measures are implemented, the ef-fects are monitored and evaluated, and supple-mentary measures are implemented if, as in thepresent case, the original targets are not attainedas expected.
Future revisions of the Danish National AquaticMonitoring and Assessment Program will be great-ly influenced and guided by adoption of the Eu-ropean Union’s Water Framework Directive (WFD;European Union 2001). The legislation must beimplemented in national law before December 22,2003 by the member countries. The monitoringobligations according to Annex V of the Directiveshould be fully implemented and fully operational
Danish Monitoring and Assessment Program 859
Fig. 7. Long-term data for Secchi disk depth and areal pri-mary production in Danish estuaries between 1977 and 1999.The line indicates the long-term mean.
by December 22, 2006. One of the many importantgoals of the WFD is to classify the state of a partic-ular environment and identify measures needed toreach stated environmental goals. Goal settingshould be an integral part of a successful monitor-ing program as exemplified by the Chesapeake BayProgram, where goals have been defined for a longlist of parameters. The goals set for the Danish Ac-tion Plan were originally defined only in terms ofreduction in loads relative to the situation in 1987without regard to the state of the environment.Those numbers were set somewhat arbitrarily in asituation where there was an immediate need forpolitical action due to severe hypoxia and anoxiain many areas. The lack of quantitative goals andoperational goals for ecological quality is probablyone of the major weaknesses of the Danish Nation-al Aquatic Monitoring and Assessment Program(Markager 2001). In the coming years with imple-
mentation of the WFD, operational goals have tobe formulated that relate to the state of the envi-ronment. Monitoring data will be essential in de-fining quality classes and quality criteria in theWFD. The EU Habitat Directive will also set con-servation goals that are relevant to marine areas.The scientific challenge for the monitoring pro-gram will be to define operational definitions forthe state of the marine environment and to estab-lish quantitative relationships between load andthe state of the environment (Markager 2001).
It is our belief that more time and money needsto be allocated for assessment activities in the cur-rent Danish National Aquatic Monitoring and As-sessment Program. The goal of the monitoringprograms are to determine the state of the marineenvironment, determine temporal trends in themarine environment, and document the effects ofthe Action Plans and other relevant measures onthe marine environment. It is implicit that withinthis framework lies knowledge about the marineenvironment (Markager 2001). Data must be pro-cessed to provide information and then further an-alyzed to gain knowledge about how systems op-erate. This level of data analysis of the Danish ma-rine monitoring data has not been conducted withsufficient strength, with the consequence that ourunderstanding of the interactions between loads ofnutrients and the state of environment is less de-veloped than possible from the existing data.
In connection with the Action Plan on theAquatic Environment, a number of concurrent re-search programs were established. The first pro-gram, the NPo Program ran from 1985 to 1990with the marine part amounting to 1.2 million C5 .The follow-up program, the Marine Research Pro-gram (HAV90) in Denmark was a significant re-search program targeted for the marine environ-ment in 1990–1995 for 10.8 million C5 . The prod-ucts from the program included a series of scien-tific reports and 300 peer reviewed publicationsincluding those listed earlier. The Strategic Envi-ronmental Program funded an aquatic sectionfrom 1995–1998 and, in addition to publishingtheir findings in the literature, the project finishedwith an undergraduate textbook about the marineenvironment (Lomstein 1999). These researchprograms have helped propel Denmark into aleading role in research in marine ecology, con-tributed to an understanding of the Danish marineenvironment, and provided critical support for theimproved management of marine areas. Prioritiesfor funding scientific research in Denmark havesince changed. These programs have been com-pleted with no new directed research money forth-coming regarding research in the marine environ-ment. The European Union has recognized the
860 D. Conley et al.
importance of using the wealth of data collectedwithin national monitoring programs, and re-search funding within the European Union (http://www.cordis.lu) has encouraged the utilization ofmonitoring data within research programs.
The Danish National Aquatic Monitoring andAssessment Program probably holds one of thebest databases in the world for evaluation of theeffects of nutrient loading on the marine environ-ment. The program has been criticized within Den-mark for not being cost-effective, not providingsufficient data for calibration of modeling efforts,and not including enough modeling efforts intothe data analysis (Dahl-Madsen 2000). Some of thecriticisms were valid (Riemann et al. 2001), andmodeling efforts are currently being directed atseveral levels (empirical models, simple budgets,and more advanced models). The recent experi-ence from the Chesapeake Bay Program Water-Quality Model review of the three-dimensionalcoupled hydrodynamic-water quality model (Sci-entific and Technical Advisory Committee 2000)highlights the danger of relying on only one modelin order to develop management alternatives.
The Danish National Aquatic Monitoring andAssessment Program has been modified since itsinception based on internal and external evalua-tions and reflects compromises among politicalgoals, government institutions, local authorities,and available funding. Modifications will occurwith the next phase planned to run from 2004–2009. While monitoring programs by their natureare essentially conservative, in order to maintainlong time series essential to document long-termchanges, the program must also have the flexibilityto adapt to changing needs. Future revisions of theDanish National Aquatic Monitoring and Assess-ment Program will encounter increased responsi-bilities especially regarding European Union leg-islation and the Water Framework Directive andthe Habitat Directive, and will need to adapt toproposed decreases in funding levels and the ad-dition of emerging technologies, such as ships-of-opportunity and instrument buoys.
ACKNOWLEDGMENTS
We express our thanks to Jørn Kirkegaard and Tony Chris-tensen, Danish Environmental Protection Agency and Kitt BellAndersen, Department of Agriculture and Biotechnology, Dan-ish Forest and Nature Agency.
LITERATURE CITED
ANDERSEN, J., L. M. MUNK, AND S. PEDERSEN. 2001. Vandram-medirektivet, indhold og perspektiver. Vand og Jord 8:17–21.
BILLEN, G., C. LANCELOT, AND M. MEYBECK. 1991. N, P, and Siretention along the aquatic continuum from land to ocean,p. 19–44. In R. F. C. Mantoura, J.-M. Martin, and R. Wollast
(eds.). Ocean Margin Processes in Global Change. John Wiley& Sons Ltd., Chichester, U.K.
BRANDT, J., J. H. CHRISTENSEN, L. M. FROHN, R. BERKOWICZ, AND
F. PALMGREN. 2000. The DMU-ATMI THOR air pollution fore-cast system—System description. Report No. 321. Ministry ofthe Environment and Energy, National Environmental Re-search Institute, Roskilde, Denmark.
CHRISTENSEN, P. B., F. MØHLENBERG, L. C. LUND-HANSEN, J. BOR-UM, C. CHRISTENSEN, S. E. LARSEN, M. E. HANSEN, J. ANDERSEN,AND J. KIRKEGAARD. 1998. The Danish marine environment:Has action improved its state? Havforskning fra MiljøstyrelsenNo. 62. Danish Environmental Protection Agency, Copenha-gen, Denmark.
CONLEY, D. J. AND A. B. JOSEFSON. 2001. Hypoxia, nutrient man-agement and restoration in Danish waters, p. 425–434. In N.N. Rabalais and R. E. Turner (eds.). Coastal Hypoxia: Con-sequences for Living Resources and Ecosystems. Coastal andEstuarine Studies No. 58. American Geophysical Union,Washington, D.C.
CONLEY, D. J., H. KAAS, F. MØHLENBERG, B. RASMUSSEN, AND J.WINDOLF. 2000. Characteristics of Danish estuaries. Estuaries23:820–837.
DAHL-MADSEN, K. I. 2000. Overvagning af vandmiljøet i Dan-mark—kan det ikke gøres lidt bedre? Vand og Jord 7:109–110.
DANISH ENVIRONMENTAL PROTECTION AGENCY. 1991. Environ-mental impacts of nutrient emissions in Denmark. Re-degørelse fra Miljøstyrelsen No. 1. Danish Environmental Pro-tection Agency, Copenhagen, Denmark.
DANISH ENVIRONMENTAL PROTECTION AGENCY. 1995. EsbjergDeclaration. 4th International Conference on the Protectionof the North Sea, Esbjerg, Denmark, 8–9 June 1995. Ministryof Environment and Energy. Danish Environmental Protec-tion Agency, Copenhagen, Denmark.
DANISH ENVIRONMENTAL PROTECTION AGENCY. 2000. Aquatic En-vironment 1999. Redegørelse fra Miljøstyrelsen No. 3. DanishEnvironmental Protection Agency, Copenhagen, Denmark.
ELLERMANN, T., O. HERTEL, AND C. A. SKJØTH. 2000. Atmosfæriskdeposition 1999, NOVA 2003. Report No. 332. Ministry of theEnvironment and Energy, National Environmental ResearchInstitute, Roskilde, Denmark.
EUROPEAN UNION. 2001. Directive 2000/60/EC of the EuropeanParliament and of the Council of 23 October 2000 establish-ing a framework for community action in the field of waterpolicy. Official Journal of the European Communities, 43:L327:1–72, Brussels, Belgium.
FALLESEN, G., F. ANDERSEN, AND B. LARSEN. 2000. Life, death andrevival of the hypertrophic Mariager Fjord, Denmark. Journalof Marine Systems 25:313–321.
FOLKETINGET. 1997. Handlingsplan mod forurening af det dan-ske vandmiljø med næringssalte. Folketinget 1986–87. Bladnr. 1100, Copenhagen, Denmark.
HELCOM. 1992. Convention on the protection of the marineenvironment of the Baltic Sea area. Helsinki Commission,Helsinki, Finland.
HERTEL, O., J. CHRISTENSEN, E. H. RUNGE, W. A. H. ASMAN, R.BERKOWICZ, M. F. HOVMAND, AND Ø. HOV. 1995. Developmentand testing of a new variable scale air pollution model—AC-DEP. Atmospheric Environment 29:1267–1290.
HOLMBOE, N., H. JENSEN, AND F. Ø. ANDERSEN. 1999. Nutrientaddition bioassays as indicators of nutrient limitation of phy-toplankton in an eutrophic estuary. Marine Ecology Progress Se-ries 186:95–104.
IVERSEN, T. M., R. GRANT, AND K. NIELSEN. 1998. Nitrogen en-richment of European inland and marine waters with specialattention to Danish policy measures. Environmental Pollution102:771–780.
JØRGENSEN, B. B. AND K. RICHARDSON (EDS.). 1996. Eutrophica-tion in coastal marine ecosystems. Coastal and Estuarine Stud-
Danish Monitoring and Assessment Program 861
ies Series, No. 52. American Geophysical Union, Washington,D.C.
JOSEFSON, A. B. AND B. RASMUSSEN. 2000. Nutrient retention inbenthic biomass of Danish estuaries: Importance of nutrientload and residence time. Estuarine Coastal and Shelf Science 50:205–216.
KAAS, H., Ø. MOESTRUP, J. LARSEN, AND P. HENRIKSEN. 1999. Gif-tige alger og algeopblomstringer. Ministry of the Environ-ment and Energy, National Environmental Research Institute,Roskilde, Denmark.
KRONVANG, B., G. ÆRTEBJERG, R. GRANT, P. KRISTENSEN, M. HOV-MAND, AND J. KIRKEGAARD. 1993. Nationwide monitoring ofnutrients and their ecological effects: State of the Danishaquatic environment. Ambio 22:176–187.
LOMSTEIN, B. (ED.). 1999. Havmiljøet ved artusindskiftet. Olsen& Olsen, Fredensborg, Denmark.
MARKAGER, S. 2001. Miljøovervagning og forskning—To sider afsamme sag. Vand & Jord 8:76–78.
MARKAGER, S., T. G. NIELSEN, J. CARSTENSEN, D. CONLEY, K.DAHL, J. HANSEN, P. HENRIKSEN, A. JOSEFSON, M. LARSEN, B.PEDERSEN, B. RASMUSSEN, J. STRAND, G. ÆRTEBJERG, H. FOSS-ING, J. S. LAUERSEN, O. HERTEL, H. SKOV, L. M. SVENDSEN, M.CLEEMANN, AND G. PRITZL. 1999. Marine omrader—Statusover miljøtilstand i 1998. Report No. 290. Ministry of the En-vironment and Energy, National Environmental Research In-stitute, Roskilde, Denmark.
MIDDLEBOE, A. L. AND K. SAND-JENSEN. 2000. Long-term changesin macroalgal communities in a Danish estuary. Phycologia 39:245–257.
MIDDLEBOE, A. L., K. SAND-JENSEN, AND D. KRAUSE-JENSEN. 1998.Patterns of macroalgal species diversity in Danish estuaries.Estuaries 34:457–466.
MINISTRY OF AGRICULTURE. 1991. Handlingsplan for bæredygtigudvikling i landbruget. Ministry of Agriculture, Copenhagen,Denmark.
MINISTRY OF ENVIRONMENT AND ENERGY. 1998. Lov ommiljøbeskyttelse. Lovbekendtgørelse nr. 698 af 22. September1998. Ministry of Environment and Energy, Copenhagen,Denmark.
MINISTRY OF ENVIRONMENT AND ENERGY. 1999. Natur- ogMiljøpolitisk Redegørelse. Ministry of Environment and En-ergy, Copenhagen, Denmark.
OPHELIA. 1995a. International Journal of Marine Biology, Vol-ume 41.
OPHELIA. 1995b. International Journal of Marine Biology, Vol-ume 42.
OSPAR. 1992. Convention for the Protection of the Marine En-vironment of the North-East Atlantic. Oslo-Paris Commission,London.
PAABY, H. AND F. MØHLENBERG. 1996. Kvælstofbelastning af ha-vmiljøet. Danish Ministry of the Environment and Energy, Na-tional Environmental Research Institute, Thematic Report1996/9, Roskilde, Denmark.
RASK, N., S. E. PEDERSEN, AND M. H. JENSEN. 1999. Response tolowered nutrient discharges in the coastal waters around theisland of Funen, Denmark. Hydrobiologia 393:69–81.
RIEMANN, B., S. MARKAGER, AND J. CARSTENSEN. 2001. Overagn-ing af vandmiljøet I Danmark—Det kan gøres bedre. Vand ogJord 7:122–124.
SAND-JENSEN, K., S. L. NIELSEN, J. BORUM, AND O. GEERTZ-HAN-SEN. 1994. Phytoplankton and macrophyte development inthe Danish coast zone. Havforskning fra Miljøstyrelsen No. 30.Danish Environmental Protection Agency, Copenhagen, Den-mark.
SCIENTIFIC AND TECHNICAL ADVISORY COMMITTEE. 2000. Reviewof the Water Quality Model. Chesapeake Bay Program Office,Annapolis, Maryland.
SVENDSEN, L. M., B. KRONGVANG, P. KRISTENSEN, AND P.GRÆSBØLL. 1995. Dynamics of phosphorus compounds in alowland river system: Importance of retention and non-pointsources. Hydrological Processes 9:119–142.
SVENDSEN, L. M., B. KRONGVANG, A. R. LAUBEL, S. E. LARSEN,AND B. ANDERSEN. 1998. Phosphorus retention in a Danishlowland river system. Verhandlungen Internationle Vereinigung fur.Limnologie 26:956–962.
SOURCES OF UNPUBLISHED MATERIALS
CHRISTENSEN, T. Personal communication. Danish Environmen-tal Protection Agency, 29 Strandgade, Copenhagen, Den-mark.
Received for consideration, July 6, 2001Accepted for publication, February 19, 2002
www.elsevier.com/locate/marpolbul
Marine Pollution Bulletin 49 (2004) 283–290
Viewpoint
Palaeoecology, reference conditions and classification ofecological status: the EU Water Framework Directive in practice
Jesper H. Andersen a,*, Daniel J. Conley a,b, Søren Hedal c
a Department of Marine Ecology, National Environmental Research Institute, Frederiksborgvej 399, P.O. Box 358, DK-4000 Roskilde, Denmarkb Department of Marine Ecology, University of Aarhus, Finlandsgade 14, DK-8200 Aarhus, Denmark
c Roskilde County, Aquatic Environment Division, Køgevej 80, DK-4000 Roskilde, Denmark
Abstract
The European Union’s Water Framework Directive (WFD) requires that all Member States within the European Union
determine reference conditions for aquatic ecosystems to provide a baseline against which to measure the effects of past and present
activities. Reference conditions are subsequently used to classify the ecological status of European waters. The decisions regarding
environmental status will be important future elements in the management of European coastal waters. We have developed a
number of classification scenarios for total nitrogen (TN) in the overlying waters of the southern part of Roskilde Fjord, Denmark,
taking as our basis a palaeoecological reconstruction of fluctuations in TN between 1850 and 1995. We present a provisional
classification scheme for the ecological status of Roskilde Fjord, sensu the WFD. Decision(s) regarding the deviation from reference
conditions will give a wide range of apparent ecological status from good, through moderate and poor, to bad depending upon the
definition of an acceptable deviation from reference conditions. The determination of an acceptable deviation will ultimately be a
political decision, and will result in a wide range in the protection of coastal waters in Europe. There is still, however, an urgent need
for a sound scientific documentation of the various scenarios for the implementation of the WFD.
� 2004 Elsevier Ltd. All rights reserved.
Keywords: Eutrophication; Nitrogen; Palaeoecology; Reference conditions; Classification; The EU Water Framework Directive
1. Introduction
For more than 30 years nutrient enrichment has been
considered a major threat to the health of coastal marine
waters (Ryther and Dunstan, 1971; Nixon, 1995; Elm-gren, 2001, de Jonge et al., 2002). Many national and
international initiatives have been implemented in order
to reduce the inputs and effects of nutrients in coastal
waters, e.g. Conley et al. (2002). Within Europe, the
most recent legislation is the European Union’s Water
Framework Directive (Anon., 2000; Elliot et al., 1999).
The Water Framework Directive (WFD) provides
a framework for the protection of ground water, in-land surface waters, transitional waters (estuaries) and
coastal waters (Anon., 2000). The overall aim of the
WFD is (1) to prevent further deterioration, protect and
enhance the environmental status of aquatic systems,
*Corresponding author. Tel.: +45-4630-1280; fax: +45-4630-1211.
E-mail address: [email protected] (J.H. Andersen).
0025-326X/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.marpolbul.2004.04.014
and (2) to promote the sustainable use of water, while
progressively reducing or eliminating discharges, losses
and emissions of pollutants and other pressures for the
long-term protection and enhancement of the aquatic
environment. The WFD provides national and localauthorities with a legislative basis for the maintenance
and recovery of water quality to achieve good ecological
and chemical status for all surface waters and good
chemical status for groundwater. Accordingly, the WFD
is considered to be the most significant piece of legisla-
tion regarding water policy produced in the last 20
years.
The coastal waters covered by the WFD with respectto biological features are generally limited to surface
waters one nautical mile from the coast. With respect to
chemical features the areas in question are limited to the
surface waters 12 nautical miles from the coast. Open
marine waters are not covered, but the WFD is likely to
influence management of all marine ecosystems because
all land-based inputs of pollutants pass through the
coastal zone to the open waters.
284 J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290
The WFD requires EU Member States to develop
classification systems to describe the ecological status of
a given water body at a given time. We use here a pal-
aeoecological method to establish reference conditionsfor TN concentration in a given coastal water (Clarke
et al., 2003). Using these reference conditions we present
examples of how the ecological status of the southern
part of Roskilde Fjord, Denmark, could be classified
according to the principles of the WFD. In this view-
point, we show that a wide range of environmental
protection will be achieved based upon decision(s)
regarding the deviation from reference conditions andour objective is to promote a general discussion of
acceptable deviations from reference conditions.
The WFD is not the only directive seeking to improve
the eutrophication status of European coastal waters
and there are other directives seeking to manage both
inputs and the ecological structure and functioning of
coastal ecosystems (Elliot et al., 1999; Elliot and de
Jonge, 2002). Inputs of nutrients from point sources andlosses of nitrogen from agricultural practices are man-
aged via the Urban Waste Water Directive and the
Nitrates Directive, respectively (Anon., 1991a,b). Other
directives, e.g. the Habitats Directive (Anon., 1992) and
the Birds Protection Directive (Anon., 1979), indirectly
influence management practices via conservation
objectives (i.e. abundance of species or food availabil-
ity), which are influenced by nutrient enrichment andeutrophication. The objectives of these directives are
summarised in Table 1.
Table 1
Summary of EU Directives other that the WFD of relevance to managemen
Directive 91/676/EEC of 12 December 1991 concerning the protection of wa
The objective of the Nitrates Directive is to reduce water pollution caused
such pollution. EU Member States shall designate vulnerable zones, which
contribute to pollution. Member States shall set up where necessary actio
agricultural practices. Member States shall also monitor and assess the eutr
years.
Directive 91/271/EEC of 21 May 1991 concerning urban waste water treatm
The objective of the Urban Waster Water Directive is to protect the envi
directive concerns the collection, treatment and discharge of urban waste
industrial sectors. The degree of treatment (i.e. emission standards) of disc
Member States shall identify areas, which are sensitive in terms of eutrop
subject to discharges.
Directive 92/43/EEC of 21 May 1992 on the conservation of natural habita
The objective of the Habitats Directive is to contribute towards ensuring bio
and fauna in European territory of the Member States. Measures shall be
natural habitats and species of wild flora and fauna of interest. The habit
respectively. Many coastal waters in the Northern Europe are identified as
implement management plans in order to restore these coastal waters and
monitor habitats and species with particular regard to priority habitat typ
Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds
The objective of the Birds Protection Directive is the long-term protection
European territory of the Member States. The directive seeks to protect, m
Member States must conserve, maintain or restore the biotopes and habit
eutrophication Member States are required to restore destroyed or impair
2. Determination of reference conditions
Establishment of reference conditions in aquatic
systems can be made in a number of different ways forthe WFD. The methods described in the legislation for
the WFD are (1) spatially based reference conditions
(including historical data), (2) modelling (empirical or
dynamic), (3) combinations of (1) and (2), and (4) expert
judgement.
If undisturbed or minimally disturbed sites are
available, reference conditions can readily be defined by
comparing other sites to these undisturbed sites. How-ever, in marine environments, where waters are easily
transported from one area to another through currents
and mixing, isolated or undisturbed sites are generally
not available, especially in the nutrient impacted en-
closed seas and coastal areas of Europe. Therefore,
alternative methods must be used to define reference
conditions in coastal waters.
The first and most widely used method to determinereference conditions is the use of monitoring data
combined with historical data on the system to estimate
conditions prior to large-scale disturbance. While his-
torical data are not abundant, there is a surprising
amount of data collected from earlier surveys that can
be mined and used to help estimate reference conditions.
Another important methodology that is widely used
to estimate reference conditions is to make predictivewatershed models to estimate long-term trends in
nutrient loading with changes in land-use and agricul-
t of coastal eutrophication
ters against pollution caused by nitrates from agriculture
or induced by nitrates from agricultural sources and to prevent further
are areas of land draining into waters affected by pollution and which
n programmes promoting the application of the codes of good
ophication status of freshwater, estuaries and coastal waters every four
ent
ronment from the adverse effects of discharges of waster-water. The
water and the treatment of discharges of waste water from certain
harges is based on assessment of the sensitivity of the receiving waters.
hication. Competent authorities shall monitor discharges and waters
ts and of wild fauna and flora
diversity through the conservation of natural habitats and of wild flora
designed to maintain or restore, at favourable conservation status,
ats and species protected are identified and defined in Annex I and II,
eutrophic due to anthropogenic inputs. Member States are required to
to achieve a favourable conservation status. Member States shall
es and priority species.
and conservation of all birds naturally living in the wild within the
anage and regulated all bird species, including eggs, nests and habitats.
ats of these birds. As many birds lives in coastal waters subject to
ed biotopes and to maintain undisturbed habitats.
J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290 285
tural practices (Billen and Garnier, 1997). These models
can also be used to estimate the effects of changes in
management practice on improvements in water quality.
An alternative methodology is the use of palaeo-reconstruction using relationships between fossil re-
mains and environment to infer past conditions (ter
Braak and Juggins, 1993). Palaeoecological methods
based on quantitative transfer functions have received
widespread use in freshwaters (e.g. Bennion et al., 1996),
but have only recently been applied to coastal waters
(Clarke et al., 2003; Weckstr€om et al., in press). All
methodologies require the use of expert judgement andthe final determination of reference conditions benefits
from a combination of methodologies.
A quantitative palaeoecological inference-model (a
weighted averaging-partial least squares diatom-based
transfer function) has recently been developed for
Danish coastal waters (Clarke et al., 2003). The transfer
function can be used to estimate historical changes in
TN concentrations in overlying waters from the bio-logical record of environmental change, preserved in
fine-grained coastal sediments as fossil diatom assem-
blages. Clarke et al. (2003) describe the methodology
and present diatom-inferred TN estimates from a 1 m
sediment core from the southern part of Roskilde Fjord
(location: 55�40092N and 11�58009E) that represents thetime period from �1850. The reconstruction by Clarke
et al. (2003) indicated that historical annual mean TNconcentrations fluctuated between 58 lmol l�1 (�1850)and 50 lmol l�1 (�1950). A rapid increase in annual
mean TN concentration is seen after the mid-1950s
with the highest diatom-inferred TN concentration (91
lmol l�1) corresponding to the surface sample collected
in 1995.
The palaeo-reconstruction by Clarke et al. (2003) can
be compared to other methodologies used to establishreference conditions in Denmark. The 1983 Manage-
ment Plan for Roskilde Fjord and its catchment esti-
mated annual mean TN concentrations in reference
conditions to be on the order of 60–65 lmol l�1 using
modelling (Hovedstadsr�adet, 1983). By comparison, in
Odense Fjord, a relatively small (65 km2) and nutrient-
impacted Danish estuary (Conley et al., 2000), reference
conditions in terms of annual TN means have beenestimated to 48–51 and 19–23 lmol l�1 for a station in
the inner estuary (55�27011N, 10�28069E) and one in the
middle of the estuary (55�28075N, 10�31015E), respec-tively, also using a modelling approach (DHI, 2002,
2003) that combines hydrodynamic transport and bio-
logical features in a 3-D fashion (MIKE 3 model). The
actual numbers have been estimated by using the cli-
matic and meteorological conditions in 1998 and 2002.However the scenarios radically altered various input
data (e.g. nutrient loads and concentrations in rivers,
atmosphere and the connecting coastal sea), and process
specifications (e.g. growth characteristics of phyto-
plankton and macroalgae) to fit into a so-called ‘‘natural
state’’ concept from which the reference conditions have
been drawn. In a third example from Denmark, Randers
Fjord, the reference concentrations of TN have beenestimated to be �57 lmol l�1, using dynamic, deter-
ministic modelling (Nielsen et al., 2003).
On a cautionary note, we know that European estu-
arine and coastal systems have experienced massive
changes in nutrient loading from anthropogenic activi-
ties during the history of human occupation (Billen and
Garnier, 1997). Increases in nutrient loading between
pristine conditions to present are certainly much largerthan those that have occurred during the last 100–150
years (Conley, 1999). For example, a historical recon-
struction of P concentrations in Dallund Lake on the
island of Fyn, Denmark has shown that the largest
changes in P concentrations occurred during the Middle
Ages associated with changes in land-use, e.g., forest
clearance, increasing in concentration 6–8 times from
the year 1000 to 1200 AD (Bradshaw, 2001). Therefore,reference conditions determined for the time prior to the
intensification of agriculture will not necessarily reflect
an unimpacted state.
Variations in TN concentration through time are
observed in the palaeoecological reconstruction in
Roskilde Fjord, Denmark, by Clarke et al. (2003), with
the lowest nutrient concentrations observed in the 1950s.
Changes in nutrient loading with changing land usepractices (Billen and Garnier, 1997), industrialisation
(Billen et al., 1999), conversion from animal derived
locomotion to the combustion engine (Nixon, 1997),
development of sewage treatment practices, and climate,
combine to influence nutrient concentrations. In addi-
tion, estuarine systems are dynamic entities where
freshwater and associated nutrients from the land are
mixed with water from coastal waters, such that theaverage nutrient concentration will be to a great extent
dependent upon location in the estuary.
3. Classification of ecological status
The WFD requires classification, in terms of ecolog-
ical status, for all European surface waters. The classi-fication should be based on reference conditions, which
are intended to represent minimal anthropogenic im-
pact, and observed deviation from these conditions.
Ecological status is to be expressed as a numerical value
(the ecological quality ratio) between 1 (high ecological
status) and 0 (bad ecological status) with intervals
equating to: high, good, moderate, poor and bad eco-
logical status (Fig. 1). The lower value in the goodecological status bracket is equivalent to the higher
value in the moderate ecological status bracket and has
been set as the management quality standard. The WFD
requires that all waters must be restored to above this
Fig. 1. Basic principles for the classification of ecological status based on ecological quality ratios (adapted from Anon., 2003). In this classification,
equal intervals between the different classes of ecological status are assumed.
286 J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290
value by 2017 (Anon., 2000). The ecological quality
ratio is defined as the relation between the observed
condition and the reference condition. The management
quality standard is described as an ‘‘ecological status
that shows a low level of distortion resulting from hu-
man activity, but deviates only slightly from those nor-
mally associated with the water body type under
undisturbed conditions’’ (Anon., 2000). The legislationdefines these conditions to be pristine with no or very
minor deviations from undisturbed conditions. Practi-
cally they are being defined as conditions prior to the
intensification of agriculture 100–150 years ago as the
post-war intensification of agriculture and urban pollu-
tion are believed to have had the largest impacts on
coastal waters. Regarding nutrients it means ‘‘concen-
trations do not exceed the levels established so as toensure the functioning of the ecosystem and the
achievement of the values specified for biological quality
elements’’ (Anon., 2000).
The classification of ecological status sensu the WFD
and the management quality standard of ‘‘good eco-
logical status’’ is likely to match the sensitive (and less
sensitive) areas sensu the Urban Waste Water Treatment
Directive and the vulnerable (and less vulnerable) zonessensu the Nitrates Directive (Fig. 2). Correspondence of
management practices and standards between directives
is important and a situation with different objectives and
standards is avoided. If the correspondence is not
Fig. 2. Proposed correspondence between eutrophication related ecologica
Directive (WFD) and other direct or indirect related EC Directives (Urban W
Protection Directive), the OSPAR Common Procedure and the present Dan
identical, management of coastal eutrophication could
become complicated beyond reason.
We consider the estimated annual mean TN concen-
trations fromClarke et al. (2003) using a palaeoecological
methodology to provide an independent, robust estimate
of past nutrient concentrations consistent with other
independent methods of estimation. Themethod could be
used to determine reference condition(s), these being thebest condition(s) within ‘‘high ecological status’’.
The selection of class boundaries will be one of the
most politically sensitive issues in setting ecological
quality standards. Once a reference value is decided, a
deviation from this value must be set in order to deter-
mine the border between a high and a good ecological
status. It is expected that a suitable boundary will be
assigned from evaluation of data or historical records(Anon., 2003). The critical value will be the border be-
tween good and moderate ecological status. Since the
WFD requires that all waters be restored to be above
this value, the threshold value between good and mod-
erate status will have significant economic consequences.
The remaining class boundaries below this value will be
set to account for the remainder of the scale. Initially
member states may set their own class boundaries,however harmonisation into a common scale will be
made within Europe through intercalibration exercises.
At present there is no commonly accepted pan-
European guidance on what an acceptable deviation
l quality objectives (EQOs) according to the EU Water Framework
aste Water Directive, Nitrates Directive, Habitats Directive and Birds
ish quality objectives. Modified from Ærtebjerg et al. (2003).
J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290 287
from reference conditions is and it is unclear what level
of deviation is to be deemed slight or moderate. To our
knowledge no country in Europe has at this stage tried
to ‘‘translate’’ the WFD lingo of what ‘‘no’’, ‘‘slight’’ or‘‘moderate’’ deviation from reference conditions are into
a meaningful classification scheme based on reference
conditions. We believe that this paper is the first attempt
to do so and we emphasise that this Viewpoint aims to
open or facilitate the discussion, not to prejudge the
outcome. Classification of ecological status for the
WFD and the development of classification schemes for
eutrophic European coastal waters appear to be in anearly stage (SFT, 1997; Swedish EPA, 2000; OSPAR,
2001). Further development is needed to meet the
requirements of the WFD.
To our knowledge, only a few classifications of
European coastal waters based on reference conditions
have been published (Krause-Jensen et al., in press).
Most existing Danish eutrophication assessment criteria
have defined a 25% deviation from reference conditionsas being acceptable (Ærtebjerg et al., 2003). Krause-
Jensen et al. (in press) discuss acceptable deviations
from reference conditions (both 25% and 15%). These
authors demonstrate that type specific classification
sensu the WFD could result in both false-positive and
false-negative results. They suggest site-specific classifi-
cation might be an alternative.
4. Application to a Danish estuary
We have combined these reference values for TN in
the south-eastern part of Roskilde Fjord (from Clarke
et al., 2003) with the Danish management suggestions
Table 2
Draft classification scenarios (A–C) of ecological status of TN in seawater i
Scenario Acceptable
deviation (%)
Reference condition
(lmol l�1)
High
(100–90%)
A 25 58 58–62
B 25 54 54–59
C 25 50 50–55
The acceptable deviation from the reference condition (being the value definin
to be 25% of the reference value. The vertical line indicates this managem
conditions in lmol l�1.
Table 3
Draft classification scenarios (D–F) of ecological status of TN in seawater i
Scenario Acceptable
deviation (%)
Reference conditions
(lmol l�1)
High
(100–95%)
D 15 58 58–61
E 15 54 54–57
F 15 50 50–52
The acceptable deviation from the reference condition (being the value definin
to be 15% of the reference value.
from Ærtebjerg et al. (2003) and Krause-Jensen et al. (in
press). In practice we have selected three reference
conditions representing a range in nutrient concentra-
tions (50, 54 and 58 lmol l�1) and assumed that 25% and15% (from Ærtebjerg et al., 2003; Krause-Jensen et al.,
in press) may be considered a first and provisional def-
inition of what an acceptable deviation from the refer-
ence conditions might be. This results in six
classification scenarios (Tables 2 and 3).
According to the six scenarios (A–F), the current
ecological state of the southern parts of Roskilde Fjord,
when using a 5-year weighted mean (78 lmol l�1) can becharacterised as being at best ‘‘moderate’’ or at worst as
‘‘poor’’. Focusing on a single year will change the out-
come of the assessment as shown in Fig. 3.
Which of the six scenarios (A–F) should be used for
the future management of Roskilde Fjord? The decision
will be political and will ultimately have a large impact
upon the level of protection of the marine environment.
Recently, Roskilde County Council has agreed on eco-logical objectives for submerged aquatic vegetation
(Zostera marina) (HUR, 2001). This objective is likely
to be fulfilled when the annual mean TN concentra-
tions are reduced below 57 lmol l�1 (Roskilde County,
unpublished data). This is more or less equal to the most
restrictive scenario (scenario F) with reference condi-
tions set at 50 lmol l�1 and an acceptable deviation
assumed to be 15%.The Swedish EPA has developed a comprehensive
classification system for coastal and open marine waters
(Swedish EPA, 2000). This system, which include five
status classes and uses TN concentrations in surface
water during winter and summer as indices, is not based
on reference conditions as required by the WFD, but on
n the southern Roskilde Fjord
Good
(90–75%)
Moderate
(75–55%)
Poor
(55–30%)
Bad
(30–0%)
63–72 73–84 85–99 >99
60–67 68–78 79–92 >92
56–62 63–72 73–85 >85
g the border between good and moderate ecological status) is assumed
ent standard. The acceptable deviation is expressed in %, reference
n the southern Roskilde Fjord
Good
(95–85%)
Moderate
(85–65%)
Poor
(65–35%)
Bad
(35–0%)
62–68 69–78 79–96 >96
58–62 63–73 74–89 >89
53–57 58–67 68–83 >83
g the border between good and moderate ecological status) is assumed
Fig. 3. Ecological status of the southern part of Roskilde Fjord using seasonally weighted mean TN concentrations for the period 1988–2002
according to scenarios A and F. Scenario A represents an acceptable deviation of 25% and a reference condition of 58 lmol l�1. Scenario F represents
an acceptable deviation of 15% and a reference condition of 50 lmol l�1. The size of the quality classes varies and is determined by the percent
deviation from reference conditions. The ecological quality ratio (EQR) is shown to illustrate the relation between TN and EQR.
288 J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290
available monitoring results for both open waters
and coastal waters. A direct comparison should there-
fore be done with caution, however, if the Swedish
assessment criteria are applied to Roskilde Fjord, the
TN concentrations are to be characterised as ‘‘very
high’’ equivalent to ‘‘bad ecological status’’. The same
results would occur using nutrient concentrations in theMediterranean Sea as the standard for Danish coastal
waters. A concentration determined for one location or
Member State cannot be applied on a pan European
basis.
Roskilde Fjord is located within the area of the
OSPAR Convention (see www.ospar.org) and is there-
fore also subject to the OSPAR Common Procedure for
the Identification of the Eutrophication Status of the
Maritime Area (OSPAR, 1998, 2001). In this context,
the generally acceptable deviation from background
concentrations or reference conditions is 50%. If this
general principle is applied to Roskilde Fjord, the
acceptable concentrations will be in the range of 75–87
lmol l�1, indicating that the area in question is likely tobe a ‘‘non-problem area’’ with no eutrophication effects.
This conclusion disagrees with our assessment presented
above and with the annual assessment by Roskilde
County. Parts of the OSPAR principles need a scientific
review in order to test the principles and to ensure that
the strategy to combat eutrophication is based on veri-
fied information.
J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290 289
5. Conclusions and perspectives
The definition of reference conditions and classifica-
tion of the ecological status represents an importantaspect in the implementation of the WFD. The deter-
mination by Clarke et al. (2003) of reference conditions
in a Danish estuary demonstrate that the use of palae-
oecological methods produces results with a reasonable
degree of accuracy. However the subsequent decision on
the deviation from the reference conditions actually
controls the classification of the ecological status, and
complicates an objective classification. The palaeoeco-logical approach applied here is a useful tool to recon-
struct reference conditions in Roskilde Fjord and
possibly also in other estuaries and coastal marine areas.
There is an urgent need to discuss and define proce-
dures to establish reference conditions and quality clas-
ses in a variety of environmental conditions. It is
important to translate ‘‘slight’’ and ‘‘moderate’’ devia-
tions into acceptable deviations from reference condi-tions. However, the implementation of the WFD is still
only in its beginning phase. The coming years will,
therefore, be a learning process during which our pro-
posed classification will be challenged and tested. In this
process it is important to integrate science and manage-
ment (Gray, 1999; Elliot et al., 1999) in order to ensure
that the WFD will be a strong instrument to manage the
ecological quality of European coastal waters.
Acknowledgements
Thanks to our colleagues Annemarie Clarke,
Bo Riemann, Jens Brøgger Jensen (Danish EPA) and
Michael Hjort Jensen (Fyn County) for valuable dis-
cussion and comments on earlier versions of the man-uscript. This research was partially supported by
European Union funding of the MOLTEN (contract
EVK3-CT-2000-00031) and CHARM (contract EKV3-
CT-2001-0065) projects.
References
Anon., 1979. Council Directive 79/409/EEC of 2 April 1979 on the
conservation of wild birds. Official Journal L 103.
Anon., 1991a. Council Directive of 21 May 1991 concerning urban
waste water treatment (91/271/EEC). Official Journal L 135.
Anon., 1991b. Council Directive 91/676/EEC of 12 December 1991
concerning the protection of waters against pollution caused by
nitrates from agricultural sources. Official Journal L 375.
Anon., 1992. Council Directive 92/43/EEC of 21 May 1992 on the
conservation of natural habitats and of wild fauna and flora.
Official Journal L 206.
Anon., 2000. Directive 200/60/EC of the European Parliament and of
the Council of 23 October 2000 establishing a framework for
Community action in the field of water policy. Official Journal L
327/1.
Anon., 2003. Principles and methods for establishing reference
conditions and ecological status class boundaries for inland surface
waters. Policy Summary to the REFCOND Guidance Document.
CIS Working Group 2.3.
Ærtebjerg, G., Andersen, J.H., Hansen, O.S. (Eds.), 2003. Nutrients
and Eutrophication in Danish Marine Waters. A Challenge for
Science and Management. National Environmental Research
Institute. 126 pp.
Bennion, H., Juggins, S., Anderson, N.J., 1996. Predicting epilimnetic
phosphorus concentrations using an improved diatom-based
transfer function and its application to lake eutrophication
management. Environmental Science and Technology 30, 2004–
2007.
Billen, G., Garnier, J., 1997. The Phison River plume: coastal
eutrophication in response to changes in land use and water
management in the watershed. Aquatic Microbial Ecology 13, 3–
17.
Billen, G., Garnier, J., Deligne, C., Billen, C., 1999. Estimates of early
industrial inputs of nutrients to river systems: implications for
coastal eutrophication. Science of the Total Environment 243–244,
43–52.
Bradshaw, E., 2001. Linking land and lake. The response of lake
nutrient regimes and diatoms to long-term land-use change in
Denmark. Ph.D. Thesis, Copenhagen University, Denmark.
Clarke, A., Juggins, S., Conley, D.J., 2003. A 150-year reconstruction
of the history of coastal eutrophication in Roskilde Fjord,
Denmark. Marine Pollution Bulletin 46, 1615–1629.
Conley, D.J., 1999. Biogeochemical nutrient cycles and nutrient
management strategies. Hydrobiologia 410, 87–96.
Conley, D.J., Kaas, H., Møhlenberg, F., Rasmussen, B., Windolf, J.,
2000. Characteristics of Danish estuaries. Estuaries 23, 820–837.
Conley, D.J., Markager, S., Andersen, J., Ellermann, T., Svendsen,
L.M., 2002. Coastal eutrophication and the Danish National
Aquatic Monitoring and Assessment Program. Estuaries 25, 706–
719.
deJonge, V.N., Elliot, M., Orive, E., 2002. Causes, historical develop-
ment, effects and future challenges of a common environmental
problem: eutrophication. Hydrobiologica 475–476, 1–19.
DHI, 2002. Belastningsscenarier for Odense Fjord. Teknisk notat.
DHI/Fyns Amt. (in Danish).
DHI, 2003. Naturscenarie 2002 for Odense Fjord. Teknisk notat. DHI/
Fyns Amt. (in Danish).
Elliot, M., de Jonge, V.N., 2002. The management of nutrients and
potential eutrophication in estuaries and other restricted water
bodies. Hydrobiologica 475–476, 513–524.
Elliot, M., Fernandes, T.F., de Jonge, V.N., 1999. The impact of
European Directives on estuarine and coastal science and man-
agement. Aquatic Ecology 33, 311–321.
Elmgren, R., 2001. Understanding human impact on the Baltic
ecosystem: changing view in recent decades. Ambio 30, 222–231.
Gray, J.S., 1999. Using science for better protection of the marine
environment. Marine Pollution Bulletin 39, 3–10.
Hovedstadsr�adet, 1983. Teknisk baggrundsnotat, 1 Roskilde Fjord og
opland, recipientgrundlag, kvalitetskrav og konsekvensberegninger
1982. In: Forslag til Recipientkvalitetsplan og spildevandsplan for
Roskilde Fjord og opland. Planlægningsrapport nr. 36, p. 60 (in
Danish).
HUR, 2001. Regionplan for Roskilde Amt. Redegørelse. 175 pp (in
Danish).
Krause-Jensen, D., Greve, T.M., Nielsen, K., in press. Eelgrass as a
bioindicator under the European Water Framework Directive.
Water Resources Management.
Nielsen, K., Sømod, B., Ellegaard, C., Krause-Jensen, D., 2003.
Assessing reference conditions according to the European Water
Framework Directive using modelling and analysis of historical
data: an example from Randers Fjord, Denmark. Ambio 32 (4),
287–294.
290 J.H. Andersen et al. / Marine Pollution Bulletin 49 (2004) 283–290
Nixon, S.W., 1995. Coastal marine eutrophication: a definition, social
causes, and future concerns. Ophelia 41, 199–219.
Nixon, S.W., 1997. Prehistoric nutrient inputs and productivity in
Narragansett Bay. Estuaries 20, 253–261.
OSPAR, 1998. OSPAR Strategy to Combat Eutrophication. Available
from <http://www.ospar.org/eng/html/sap/eutstrat.htm>.
OSPAR, 2001. Current status of elaborated ecological quality objec-
tives for the Greater North Sea with regard to nutrients and
eutrophication effects.
Ryther, J.H., Dunstan, W.M., 1971. Nitrogen, phosphorus, and
eutrophication in the coastal marine environment. Science 171,
1008–1013.
SFT, 1997. Klassifisering av miljøkvalitet i fjorder og kystfarvann.
Veiledning fra SFT 97:03 (in Norwegian).
Swedish EPA, 2000. Coasts and Seas. Environmental Quality Crite-
ria. Swedish Environmental Protection Agency. Report 5052.
138 pp.
ter Braak, C.J.F., Juggins, S., 1993. Weighted averaging partial least
squares regression (WA-PLS): an improved method for re-
constructing environmental variables from species assemblages.
Hydrobiologia 269–270, 485–502.
Weckstr€om, K., Juggins, S., Korhola, A., in press. Defining back-
ground nutrient concentrations for coastal waters of the Gulf of
Finland, Baltic Sea. Ambio.
HORIZONS
Coastal eutrophication: recentdevelopments in definitions andimplications for monitoring strategies
JESPER H. ANDERSEN1*, LOUISE SCHLUTER1 AND GUNNI ÆRTEBJERG2
1
DEPARTMENT OF ECOLOGY AND ENVIRONMENT, DHI WATER & ENVIRONMENT, AGERN ALLE 5, 2790 HØRSHOLM, DENMARK AND2
DEPARTMENT OF
MARINE ECOLOGY, NATIONAL ENVIRONMENTAL RESEARCH INSTITUTE, FREDERIKSBORGVEJ 399, 4000 ROSKILDE, DENMARK
*CORRESPONDING AUTHOR: [email protected]
Received September 15, 2005; accepted in principle January 12, 2006; accepted for publication March 22, 2006; published online March 29, 2006
Communicating editor: K.J. Flynn
The word ‘eutrophication’ has its root in two Greek words: ‘eu’ which means ‘well’ and ‘trope’ which
means ‘nourishment’. The modern use of the word eutrophication is related to inputs and effects of
nutrients in aquatic systems. Despite a common understanding of its causes and effects, there is no
agreed definition of coastal eutrophication. This communication aims to review recent developments in
the definitions of coastal eutrophication, all of which focus on ‘accelerated growth’, and to discuss the
implications in relation to monitoring and assessment of ecological status. It is recommended that
measurements of primary production, being a sensitive and accurate indicator of eutrophication, should
be mandatory when monitoring and assessing the ecological status of coastal waters.
INTRODUCTION
Eutrophication of coastal waters has been considered one
of the major threats to the health of marine ecosystems
for more than 30 years (Ryther and Dunstan, 1971;
Nixon, 1995; Elmgren, 2001; Bachmann et al., 2006).
The different processes and effects of coastal eutrophica-
tion are well known and documented (Cloern, 2001;
Conley et al., 2002; Ronnberg and Bonsdorff, 2004).
In 2000, the European Parliament and the Council
adopted the European Union (EU) Water Framework
Directive (WFD), which provides a framework for the
protection of groundwater, inland surface waters, transi-
tional waters (estuaries) and coastal waters (Anonymous,
2000). The overall aim of the WFD was: (i) to prevent
further deterioration, protect and enhance the environ-
mental status of aquatic systems and (ii) to promote the
sustainable use of water while progressively decreasing or
eliminating discharges, losses and emissions of pollutants
and other pressures for the long-term protection and
enhancement of the aquatic environment. The WFD
is intended to improve the ecological status, including
eutrophication status, of all European surface waters of
which many are considered to be eutrophic (European
Environment Agency, 2001, 2003). The directive provides
national and local authorities with a legislative basis for the
maintenance and recovery of water quality to achieve good
ecological and chemical status for all surface waters and
good chemical status for groundwater. Accordingly, the
directive can be considered the most significant piece of
legislation of the last 20 years, in regard to water policy not
only in Europe but also in non-European countries seeing
EU legislation as a benchmark for their own legislation.
Written responses to this article should be submitted to Kevin Flynn at [email protected] within two months of publication. For
further information, please see the Editorial ‘Horizons’ in Journal of Plankton Research, Volume 26, Number 3, Page 257.
JOURNAL OF PLANKTON RESEARCH j VOLUME 28 j NUMBER 7 j PAGES 621–628 j 2006
doi:10.1093/plankt/fbl001, available online at www.plankt.oxfordjournals.org
� The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
However, the WFD lacks a definition of eutrophica-
tion. The directive’s treatment of eutrophication is indir-
ect, with the boundary between good and moderate
ecological status being defined as an environmental man-
agement objective. For waters failing to meet the objec-
tive of at least good ecological status, the directive
requires that competent authorities establish pro-
grammes of measures and river basin management
plans to secure this status. The measures to be imple-
mented in the context of eutrophication are already
required under other existing directives, for example,
the Urban Waste Water Treatment (UWWT) Directive
(Anonymous, 1991a) and the Nitrates Directive
(Anonymous, 1991b). If these are insufficient, then the
implementation of supplementary measures is required.
The WFD thus acts as an umbrella for the UWWT
Directive and the Nitrates Directive, and as such it has
to respect the definitions of eutrophication in these
directives.
HOW IS EUTROPHICATIONDEFINED?
Within the EU, there has been a sound tradition of
focusing measures on the sources causing eutrophication
(Elliot et al., 1999; Elliot and de Jonge, 2002).
Consequently, eutrophication has been defined in rela-
tion to sources and/or sectors. For example, the
European Commission (EC) UWWT Directive defines
eutrophication as ‘the enrichment of water by nutrients,
especially nitrogen and/or phosphorus, causing an accel-
erated growth of algae and higher forms of plant life to
produce an undesirable disturbance to the balance of
organisms present in the water and to the quality of
water concerned’ (Anonymous, 1991a).
According to the EC Nitrates Directive, eutrophica-
tion is defined as ‘the enrichment of water by nitrogen
compounds causing an accelerated growth of algae and
higher forms of plant life to produce an undesirable
disturbance to the balance of organisms present in the
water and to the quality of water concerned’
(Anonymous, 1991b). The difference between the two
definitions can be explained by the focus of the Nitrates
Directive which, perhaps unsurprisingly, rests on losses of
nitrogen from agriculture.
There has been some justifiable discussion of these
definitions, in particular their focus on nutrients, and
also the need to clarify what constitutes an ‘undesirable
disturbance’ and an ‘accelerated growth’. Is ‘accelerated’
the right word to use in this context? No, accelerated,
meaning speed up, is in our opinion the wrong word and
should be replaced by ‘increased’. Nixon (Nixon, 1995)
defines eutrophication as ‘an increase in the rate of
supply of organic matter to an ecosystem’. This definition
is short and consistent with historical usage and empha-
sizes eutrophication as a process rather than a trophic
state. Nixon also notes that the increase of the supply of
organic matter to coastal systems may have various
causes, but the common factor is clearly nutrient enrich-
ment. The supply of organic matter to an ecosystem is
not restricted to pelagic primary production, even
though such an interpretation leads to a convenient
operational definition. It also includes primary produc-
tion of higher plants and benthic microalgae as well as
inputs of organic matter from adjacent waters or from
land, via rivers or point sources. Having such a broad
interpretation of the term ‘supply’ makes the definition,
despite its obvious strengths, difficult to apply in a mon-
itoring and management context.
Eutrophication and definition(s) of eutrophication are
much discussed topics as indicated above and also
pointed out by Jørgensen and Richardson (Jørgensen
and Richardson, 1996). The most common use of the
term is related to inputs of mineral nutrients, primarily
nitrogen and phosphorus, to specific waters.
Consequently, eutrophication deals with both the process
as such, the associated effects of nutrient enrichment and
natural versus cultural caused eutrophication. And as
prudently pointed out by Jørgensen and Richardson,
when we speak of eutrophication, it is anthropogenic
eutrophication that is of interest.
Within the OSPAR Convention for the Protection of
the Marine Environment of the North-East Atlantic, the
definition of eutrophication follows the above definitions
and thoughts and defines eutrophication similar to the
UWWT Directive and continues ‘and therefore refers to
the undesirable effects resulting from anthropogenic
enrichment by nutrients described in the Common
Procedure’ (OSPAR, 2003).
The implementation of the WFD has revealed the
need for a common understanding and definition of
eutrophication as well as a need for stronger coordina-
tion between directives dealing directly or indirectly with
eutrophication. The EC has initiated a process with the
aim of developing a pan-European conceptual frame-
work for eutrophication assessment in the context of all
European waters and policies. At a workshop in
September 2004, hosted by the EC and Joint Research
Centre in Ispra, Italy, draft guidance on a pan-European
framework for assessment of eutrophication was pre-
sented and discussed. The objective of the workshop
was to coordinate different activities under the EU
WFD and other eutrophication-related directives (e.g.
UWWT Directive and Nitrates Directive). The
JOURNAL OF PLANKTON RESEARCH j VOLUME 28 j NUMBER 7 j PAGES 621–628 j 2006
622
workshop concluded that a draft pan-European defini-
tion of eutrophication could use the UWWT Directive as
a starting point for further developments on the issue of
eutrophication. Taking the comments put forward at the
workshop into consideration, eutrophication can be
defined as ‘the enrichment of water by nutrients, espe-
cially nitrogen and/or phosphorus and organic matter’
(Anonymous, 2004). Work is ongoing and expected to be
reported in the spring 2006 in the form of an interim
guidance document. Revision of the guidance is planned
in 2007, following the WFD inter-calibration exercise
and some on-going activities by the conventions for the
protection of the marine environment of Baltic Sea and
the North-East Atlantic.
TOWARDS A PROCESS-ORIENTEDMONITORING AND ASSESSMENTSTRATEGY
How are member states of the EU obliged to monitor
and assess the ecological status of coastal waters?
Monitoring networks should be established to create a
coherent and comprehensive overview of ecological and
chemical status and ecological potential. The networks
should be operational by 20 December 2006 or by 1
January 2007 at the latest. Monitoring networks should
in principle be based on variables/indicators that are
indicative of the status of each relevant quality element
[biological (e.g. phytoplankton, submerged aquatic vege-
tation and invertebrate benthic fauna), hydromorpholo-
gical or physiochemical]. In addition, the networks
should permit classification of water bodies in five classes
consistent with the normative definitions of ecological
status.
In a North European perspective, there are at least
two or three important drivers for the design, execution
and reporting of monitoring activities. These are the
WFD including the WFD Common Implementation
Strategy guidance on monitoring (Anonymous, 2000,
2003a), the HELCOM COMBINE Programme (Co-
operative Monitoring in the Baltic Sea Environment)
(HELCOM, 2003) and the OSPAR Joint Assessment
and Monitoring Programme (JAMP), including the
Eutrophication Monitoring Programme, which describes
the indicators and sampling methods (OSPAR, 2004,
2005). So far, the pan-European process for development
of a conceptual framework for eutrophication assessment
has not included discussion of specific monitoring gui-
dance. This will take place at a later stage. The only
available guideline for selection of indicators is a draft
holistic checklist (Anonymous, 2004).
The requirement relating to the monitoring of pelagic
biological and chemical indicators in EU WFD,
HELCOM COMBINE, OSPAR JAMP/Coordinated
Environmental Monitoring Programme (CEMP) and
the ongoing pan-European process is summarized in
Table I. Measurements of phytoplankton species abun-
dance, composition and biomass are mandatory in most
monitoring networks. Measurements of chlorophyll a
(Chl a) and nutrients are mandatory within HELCOM
and OSPAR but considered a recommended supporting
indicator by European drivers. Measurements of primary
production are not mandatory at present.
How to assess ecological status?
The WFD requires EU member states to develop classifi-
cation systems to describe the ecological status of a given
water body at a given time. The results of the monitoring
programmes are the basis for an assessment of ecological
status of a given water body that according to the directive
will fall into one of five classes (categories): high, good,
moderate, poor or bad. The status classes high and good
are in general considered to be acceptable.
An important step in assessing ecological status is the
setting of reference condition standards with the objec-
tive of enabling the assessment of ecological quality
against these standards. Reference condition is in this
context defined as a description of the biological quality
elements that exist, or would exist, at high status, that is,
with no, or very minor, disturbance from human activ-
ities (Anonymous, 2003b).
Another important step is to define what constitutes an
acceptable deviation. An acceptable deviation sensu the
WFD is to us equivalent to high and good ecological
status, the latter defined as a status where the values of
the biological quality elements show low levels of distor-
tion resulting from human activity. An unacceptable
deviation is in our understanding equivalent to bad,
poor or moderate ecological status, where values of the
biological quality elements deviate moderately or more
from those normally associated with the coastal water
body type under undisturbed conditions sensu the WFD
definition of reference conditions.
The approach employed in the so-called OSPAR
Comprehensive Procedure (COMPP) is very pragmatic
and straightforward. On the basis of background values,
in practice identical to reference conditions, a water body
is considered an ‘Eutrophication Problem Area’ if actual
status deviates 50% or more from reference conditions
(OSPAR, 2003). It should be noted that the choice of
50% is arbitrary, not based on any scientific considera-
tions about ecological changes caused by nutrient enrich-
ment. The application of percentages lower than
J. H. ANDERSEN ETAL. j COASTAL EUTROPHICATION
623
50% has been discussed, for example, by Ærtebjerg
et al. (Ærtebjerg et al., 2003), Andersen et al. (Andersen
et al., 2004) and Krause-Jensen et al. (Krause-Jensen et al.,
2005). Recently, the OSPAR Eutrophication Committee
amended the procedures of the next application of
the Comprehensive Procedure, so that the acceptable
deviation should be justified but not exceed 50%
(OSPAR, 2005).
How can primary production be estimated?
With the development in relation to a pan-European
definition of eutrophication, it would be logical to focus
monitoring on relevant biological indicators including
measurement of ‘increased growth’. In our understand-
ing, measurement of primary production is a relevant
indicator that can indicate if algal growth is increased.
Primary production is a fundamental ecological indi-
cator (variable), because it is a measure of the extent to
which primary energy input (solar energy) to the aqua-
tic environment is transformed into the biological/
ecological sphere. It is defined as the flux of inorganic
carbon into planktonic algae per unit time. It has signifi-
cant capability to indicate and characterize the status of a
particular water body. Primary production can conveni-
ently be measured using the so-called 14C method
(Steemann Nielsen, 1952). When adding a known quan-
tity of the radioactive isotope 14C to a water sample, the
planktonic algae will take up 14C along with ‘native’ 12C
present in water. After a short incubation period (2 h), the14C incorporated into the algal cells can be measured by
liquid scintillation counting. The total carbon uptake,
which is a good approximation of net production
(Jespersen et al., 1995), can then be calculated by:
12CO2 uptake ¼ ð14CO2 uptake=14
CO2 added�12 CO2 concentration
Primary production can either be determined as
particulate production or total production. For
Table I: Selection of relevant quality elements and indicators by WFD, HELCOM COMBINE,OSPAR COMPP and the draft holistic checklist of the pan-European conceptual framework foreutrophication assessment
Quality elements and indicators EU WFD HELCOM OSPAR pan-European
Phytoplankton
Abundance M M (R) (R)
Composition M M M (R)
Diversity M (R) (R) (R)
Biomass M M (R) R
Primary production n.i. R n.i. R
Chlorophyll a R M M R
Fluorescence n.i. R n.i. n.i
Transparency
Secchi depth R M R n.i.
Light attenuation n.i. Ma n.i. R
Turbidity R n.i. n.i. R
Color R R n.i. n.i.
Nutrients
Total P R M M R
Soluble reactive P R M M R
Total N R M M R
Nitrate + nitrite R M M R
Ammonium R M M R
Silicate n.i. M n.i. R
COMPP, Comprehensive Procedure; EU WFD, European Union Water Framework Directive; M, mandatory; R, recommended; (R), recommended
indirectly; n.i., no information. Compiled from Anonymous (Anonymous, 2003a; Anonymous, 2003b; Anonymous, 2004), HELCOM (HELCOM, 2003) and
OSPAR (OSPAR, 2005).aMandatory when primary production is measured.
JOURNAL OF PLANKTON RESEARCH j VOLUME 28 j NUMBER 7 j PAGES 621–628 j 2006
624
particulate production, only the 14C uptake in the
algae cells is determined, whereas total production
also includes the 14C incorporated into the organic
matter, which can be lost to the environment outside
the cell during incubation. The method is very sen-
sitive, and primary production is a widely used
method when assessing eutrophication effects in
coastal waters (e.g. Pinckney et al., 1999; Prins
et al., 1999; Bonsdorff et al., 2002). Primary produc-
tion is also used as an important indicator when
modelling how changes in loads impact upon the
environment.
Various research activities and monitoring networks
have made use of the 14C method and have docu-
mented considerable changes in the levels of the pri-
mary production since the 1950s (e.g. Richardson and
Heilmann, 1995; Bonsdorff et al., 2002). In the central
Great Belt, Denmark (558 220 3600 N, 118000 E), the
annual primary production, averaged over each dec-
ade, has roughly doubled from the 1950s to the 1980s
and 1990s (Fig. 1). In the central Kattegat, the aver-
age monthly primary production at four different
depths in the water column through the year is com-
pared for the two periods 1954–60 and 1984–93 (Fig.
2). It can be seen that both the spring bloom and the
algal production during the summer months increased
significantly from the 1950s to 1984–93, as a conse-
quence of eutrophication (Jørgensen and Richardson,
1996).
How to link the definition with monitoringand assessment activities?
Despite positive pan-European developments in defining
eutrophication, it is still unclear what an ‘undesirable
disturbance’ is. The phrase is open to interpretation
and should be reconsidered. We suggest that an ‘undesir-
able disturbance’ in ecological terms is understood as an
‘unacceptable deviation from reference conditions’. We
realize that an ‘unacceptable deviation’ is also open to
interpretation, but the advantage is 2-fold. First, the
definition will be linked to the WFD implementation
process, and second, reference conditions sensu the
WFD will be the starting point.
We also suggest inclusion of primary production mea-
surements in monitoring systems. These should be based
on a reasonable and cost-effective approach, that is,
monitoring networks should be stratified and based on
two types of stations: (i) intensive stations/areas where
many indicators are monitored with high frequency and
(ii) mapping stations where a few indicators are moni-
tored with lower frequency. This kind of stratification has
been used in the HELCOM COMBINE Programme
(HELCOM, 2003) and in Danish National Marine
Monitoring and Assessment Programme 2003–09
(DNAMAP) (Andersen, 2005).
In our opinion, measurements of primary production
should be carried out at all intensive stations or at least
one coastal station per type of coastal water or river basin
district. Sampling frequency should be based on informa-
tion on the ecological status and take seasonal variations
at the station into account.
We also recommend that primary production mea-
surements should follow the methodology developed
within International Council for the Exploration of the
Sea (ICES) and currently described in the HELCOM
COMBINE Manual (HELCOM, 2003). However, exist-
ing time series on primary production should be contin-
ued using the original measurement method.
Primary production (g C m–2 month–1)
50
40
30
20
10
0J F M A M J J A S O N D
Month
1954–19601984–1993
Fig. 2. Primary production in the Kattegat, Denmark, through theyear as estimated by Steemann Nielsen (Steemann Nielsen, 1964) andRichardson and Heilmann (Richardson and Heilmann, 1995) [FromJørgensen and Richardson (1996). Copyright 1996 AmericanGeophysical Union. Modified by kind permission of AmericanGeophysical Union].
Primary production, g C m–2 month–1)
200
150
100
50
01950s 1960s 1970s 1980s 1990s
Decades
Fig. 1. Examples of observed changes in the primary production in thecentral Great Belt, Denmark, depicted as averages of the annual pri-mary production of the decades [unpublished data from G. Ærtebjerg,NERI, Denmark].
J. H. ANDERSEN ETAL. j COASTAL EUTROPHICATION
625
We are of the opinion that the 14C method allows
precise determination of phytoplankton production.
However, these measurements are not mandatory in mon-
itoring programmes coordinated on an international level
(e.g. HELCOM COMBINE, OSPAR JAMP and WFD
related monitoring activities). If our suggestion of includ-
ing estimates of primary production in the monitoring
programmes is followed, then these programmes will be
linked directly to both the definition and process of eutro-
phication. Other methods for determining primary pro-
duction could be employed, for example, non-isotope
method, that is, the oxygen method (Hall and Moll,
1975; Reid and Shulenberger, 1986; Olesen et al., 1999).
An indicator often used for assessment of eutrophica-
tion and as a proxy for primary productivity, nutrient
status or phytoplankton biomass is Chl a. Some caution
is recommended when using this indicator, and the infor-
mation inherent in Chl a measurements should be inter-
preted as what it is: a Chl a concentration and nothing
more, cf. Kruskopf and Flynn (Kruskopf and Flynn, 2006).
DNAMAP 2003–09, which implements the monitoring
requirements of the WFD, was designed according to a
principle stating: ‘No monitoring without Ecological
Quality Objectives, no Ecological Quality Objectives with-
out monitoring’ (Svendsen and Norup, 2005). We com-
pletely agree with this principle and present a total of nine
draft classification scenarios on the basis of percentage
deviations for the various boundaries between the classes
high, good, moderate, poor and bad (Table II). The sce-
narios are site specific (The Great Belt, Denmark) and not
directly applicable to other coastal waters. They are also
specific for the results of primary production measure-
ments and may not be applicable for other indicators. As
a cautionary note, we acknowledge that the decision on
which of the presented scenarios to implement as an
environmental management standard will be political.
CONCLUSIONS
Our mission is to propose a better definition of eutrophi-
cation and to link the definition with monitoring and
assessment systems. By understanding in ecological
terms an ‘undesirable disturbance’ as an ‘unacceptable
deviation from reference conditions’, we arrive at a defi-
nition that is consistent with the normative definitions of
moderate (and poor/bad) ecological status sensu the
WFD. Consequently, an acceptable deviation will corre-
spond to the normative definition of high and good
ecological status.
Accepting the above suggestions allows a definition of
eutrophication as ‘the enrichment of water by nutrients,
especially nitrogen and/or phosphorus and organic mat-
ter, causing an increased growth of algae and higher
forms of plant life to produce an unacceptable deviation
in structure, function and stability of organisms present
Table II: Scenarios for ecological classification in the Great Belt, Denmark using primary productionas an indicator and assuming that deviations of 15% (restrictive), 25% (intermediate) and 50%(non-restrictive) from reference conditions are acceptable deviations
Scenarios Reference conditions High (%) Good (%) Moderate (%) Poor (%) Bad (%)
Restrictive Primary production <5 5–15 15–35 35–65 >65
A1 48 <50 50–55 55–65 65–79 >79
A2 67 <70 70–77 77–90 90–111 >111
A3 86 <90 90–99 99–116 116–142 >142
Intermediate <10 10–25 25–45 45–70 >70
B1 48 <53 53–60 60–70 70–82 >82
B2 67 <74 74–92 92–97 97–114 >114
B3 86 <95 95–108 95–125 125–146 >146
Non-restrictive <20 20–50 50–70 70–90 >90
C1 48 <58 58–72 72–82 82–91 >91
C2 67 <80 80–100 80–114 114–127 >127
C3 86 <103 103–129 103–146 146–163 >163
The primary production is expressed as g C m–2 year–1. Reference conditions in scenarios A1, B1 and C1 are defined by Hansen et al. (Hansen et al.,
2003). Reference conditions in scenarios A3, B3 and C3 are defined by Ærtebjerg (unpublished data). Scenarios A2, B2 and C2, where the reference is
67 g C m–2 year–1, are an average of 48 and 86 g C m–2 year–1. The approach used for division in five quality classes is based on Andersen et al. (Andersen
et al., 2004) and Krause-Jensen et al. (Krause-Jensen et al., 2005).
JOURNAL OF PLANKTON RESEARCH j VOLUME 28 j NUMBER 7 j PAGES 621–628 j 2006
626
in the water and to the quality of water concerned,
compared to reference conditions’.
In our opinion, the proposed definition of eutrophica-
tion will lead to revision of existing monitoring strategies.
Measurement of primary production, being an indicator
of ‘increased growth’, should be mandatory in monitor-
ing networks and should consequently be included as a
monitoring or an assessment indicator in the pan-
European guidance on a conceptual framework for
eutrophication assessment.
We have raised many rhetorical questions and believe
we have answered most of the questions and by doing so
promoted the idea of having a process-oriented approach
to monitoring and assessment of coastal eutrophication.
However, one important question is still to be answered:
‘How should primary production be measure or esti-
mated?’ Such question requires thorough scientific ana-
lyses as well as coordination, otherwise the answer would
be up to individual member states meaning that there will
be only limited coordination.
The approach to be employed in setting up classifica-
tions scenarios is a topic for discussion. Our intention is
simply to present some examples of how ecological
classification scenarios could be constructed on the
basis of measurements of primary production. Further
work is needed to verify both the approach and the
scenarios. However, we consider it vital that science
and management are integrated to ensure that the
WFD will be a strong legal instrument for the protection
and, where needed, restoration of the ecological status of
European waters. Implementation of the WFD is still in
its initial phases. The coming years will, therefore, be a
learning process. Agreement on a pan-European defini-
tion of eutrophication and putting emphasis on primary
production will be a good start to this process.
ACKNOWLEDGEMENTS
The authors acknowledge discussions with and inputs
from Jens Brøgger Jensen, Bo Guttmann, Juha-Markku
Leppanen and Ciaran Murray. The manuscript was
improved from comments by the referees and the com-
municating editor. J.H.A. is partly funded by the
Helsinki Commission (HELCOM EUTRO, Project
No. 11.24/05–06).
REFERENCES
Ærtebjerg, G., Andersen, J. H. and Hansen, O. S. (eds) (2003) Nutrients
and Eutrophication in Danish Marine Waters. A Challenge for Science and
Management. National Environmental Research Institute, 126 pp.
http://eutro.dmu.dk/.
Andersen, J. H. (ed.) (2005) Marine waters. In Svendsen, L. M., Bijl, L. van
der, Boutrup, S. and Norup, B. (eds), NOVANA: Nationwide Monitoring and
Assessment Programme for the Aquatic and Terrestrial Environments. Programme
Description – Part 2. National Environmental Research Institute,
Denmark, 137 pp. NERI Technical Report No. 537. http://
www2.dmu.dk/1_Viden/2_Publikationer/3_Fagrapporter/rapporter/
FR537.pdf/.
Andersen, J. H., Conley, D. J. and Hedal, S. (2004) Palaeo-ecology,
reference conditions and classification of ecological status: The EU
Water Framework Directive in practice. Mar. Pollut. Bull., 49,
282–290.
Anonymous (1991a). Council Directive of 21 May 1991 concerning
urban waste water treatment (91/271/EEC). Official Journal L 135.
Anonymous (1991b). Council Directive 91/676/EEC of 12 December
1991 concerning the protection of waters against pollution caused by
nitrates from agricultural sources. Official Journal L 375.
Anonymous (2000) Directive 200/60/EC of the European Parliament and
of the Council of 23 October 2000 establishing a framework for
Community action in the field of water policy. Official Journal L
327/1.
Anonymous (2003a) Common Implementation Strategy for the Water
Framework Directive (2000/60/EC). Monitoring under the Water
Framework Directive. Guidance Document No. 7. Produced by
Working Group 2.7 – Monitoring, 160 pp.
Anonymous (2003b) Common Implementation Strategy for the Water
Framework Directive (2000/60/EC). Guidance on Typology,
Reference Conditions and Classification Systems for Transitional and
Coastal Waters. Produced by Working Group 2.4 – Coast, 121 pp.
Anonymous (2004) Notes of Workshop on a Conceptual Framework for
the Assessment of Eutrophication in European Waters 14–15
September 2004. JRC, Ispra, 16 pp.
Bachmann, R. W., Cloern, J. E., Heckey, R. E. et al. (eds) (2006)
Eutrophication of freshwater and marine ecosystems. Limnol.
Oceanogr., 51 (1, part 2), 351–800.
Bonsdorff, E., Ronnberg, C. and Aarnio, K. (2002) Some ecological
properties in relation to eutrophication in the Baltic Sea.
Hydrobiologica, 475/476, 371–377.
Cloern, J. (2001) Our evolving conceptual model of the coastal eutro-
phication problem. Mar. Ecol. Prog. Ser., 210, 223–253.
Conley, D. J., Markager, S., Andersen, J. et al. (2002) Coastal eutrophi-
cation and the Danish National Aquatic Monitoring and Assessment
Program. Estuaries, 25, 706–719.
EEA (2001) Eutrophication in Europe’s Coastal Waters. Topic report,
7/2001. European Environment Agency, 166 pp.
EEA (2003) Europe’s water: An indicator-based assessment. Topic report,
1/2003. European Environment Agency, 97 pp.
Elliot, M. and de Jonge, V. N. (2002) The management of nutrients and
potential eutrophication in estuaries and other restricted water
bodies. Hydrobiologica, 475/476, 513–524.
Elliot, M., Fernandes, T. F. and de Jonge, V. N. (1999) The impact of
European Directives on estuarine and coastal science and manage-
ment. Aquat. Ecol., 33, 311–321.
Elmgren, R. (2001) Understanding human impact on the Baltic ecosys-
tem: changing view in recent decades. Ambio, 30, 222–231.
Hall, C. A. S. and Moll (1975) Methods of assessing primary produc-
tivity. In Leith, H. and Whittaker, R. H. (eds), Primary Productivity of the
Biosphere. Springer, New York.
J. H. ANDERSEN ETAL. j COASTAL EUTROPHICATION
627
Hansen, I. S., Uhrenholdt, T. and Dahl-Madsen, K. I. (2003)
Miljøeffektvurdering for Havmiljøet Del 2: 3D Procesbaseret Modellering Af
Miljøtilstanden I de Abne Indre Farvande. Report from Environmental
Assessment Institute (IMV), 50 pp. (in Danish).
HELCOM (2003) Manual for Marine Monitoring in the COMBINE
Programme of HELCOM. Part C. Programme for monitoring of eutro-
phication and its effects. http://www.helcom.fi/Monas/
CombineManual2/PartC/CFrame.htm/.
Jespersen, A.-M., Søndergaard, M., Richardson, K. et al. (1995) Maling
af primærproduktion hos marint fytoplankton. Havforskning fra
Miljøstyrelsen, 55, 53 pp. (in Danish with English abstract).
Jørgensen, B. B. and Richardson, K. (eds) (1996) Eutrophication in Coastal
Marine Ecosystems. In Coastal and Estuarine Studies 52. American
Geophysical Union, Washington, DC.
Krause-Jensen, D., Greve, T. M. and Nielsen, K. (2005) Eelgrass as a
Bioindicator under the Water Framework Directive. Water Resour.
Manage., 19, 63–75.
Kruskopf, M. and Flynn, K. J. (2006) Chlorophyll content and fluores-
cence responses cannot be used to gauge reliably phytoplankton
biomass, nutrient status or growth rate. New Phytol., 169, 525–536.
Nixon, S. W. (1995) Coastal marine eutrophication: a definition, social
causes, and future concerns. Ophelia, 41, 199–219.
Olesen, M., Lundsgaard, C. and Andrushaitis, A. (1999) Influence of
nutrients and mixing on the primary production and community
respiration in the Gulf of Riga. J. Mar. Syst., 23, 127–143.
OSPAR (2003) 2003 Strategies of the OSPAR Commission for the Protection of
the Marine Environment of the North-East Atlantic. II – Eutrophication.
http://www.ospar.org/eng/html/welcome.html/.
OSPAR (2004) 2003 Strategy for a Joint Assessment and Monitoring Programme
(JAMP) (as revised by OSPAR 2004). OSPAR Commission, 2004-
17-E, 22 pp.
OSPAR (2005) Summary Record of the Meeting of the Eutrophication
Committee (EUC), 10–14 January 2005, Berlin, OSPAR
Commission, EUC 05/13/1-E, 20 pp.
Pinckney, J. L., Paerl, H. W. and Harrington, M. B. (1999)
Responses of the phytoplankton community growth rate to nutri-
ent pulses in variable estuarine environments. J. Phycol., 35,
1455–1463.
Prins, T. C., Escaravage, V., Wetsteyn, L. P. M. J. et al. (1999) Effects of
different N- and P-loading on primary and secondary production in
an experimental marine ecosystem. Aquat. Ecol., 33, 65–81.
Reid, J. L. and Shulenberger, E. (1986) Oxygen saturation and carbon
uptake near 288N, 1558W. Deep-Sea Res., 33, 271–273.
Richardson, K. and Heilmann, J. P. (1995) Primary production in the
Kattegat: Past and present. Ophelia, 41, 317–328.
Ronnberg, C. and Bonsdorff, E. (2004) Baltic Sea eutrophication: area-
specific ecological consequences. Hydrobiologia, 514, 227–241.
Ryther, J. H. and Dunstan, W. M. (1971) Nitrogen, phosphorus, and
eutrophication in the coastal marine environment. Science, 171,
1008–1013.
Steemann Nielsen, E. (1952) The use of radioactive carbon (C14) for
measuring organic production in the sea. J. Cons. Int. Explor. Mer., 18,
117–140.
Steemann Nielsen, E. (1964) Investigations of the rate of primary
production at two Danish light ships in the transition area between
the North Sea and the Baltic. Medd. Dan. Fisk. Havunders. N.S., 4,
31–77.
Svendsen, L. M. and Norup, B. (2005) NOVANA. Nationwide Monitoring
and Assessment Programme for the Aquatic and Terrestial Environments.
Programme Description – Part 1. National Environmental. Research
Institute, Denmark, 53 pp. Technical Report No. 532.
JOURNAL OF PLANKTON RESEARCH j VOLUME 28 j NUMBER 7 j PAGES 621–628 j 2006
628
398
Limnol. Oceanogr., 51(1, part 2), 2006, 398–408q 2006, by the American Society of Limnology and Oceanography, Inc.
Coastal eutrophication and trend reversal: A Danish case study
Jacob Carstensen1
Department of Marine Ecology, National Environmental Research Institute, P.O. Box 358, DK-4000 Roskilde, Denmark;European Commission, Joint Research Centre, Institute for Environment and Sustainability, TP 280, I-21020Ispra (VA), Italy
Daniel J. ConleyDepartment of Marine Ecology, National Environmental Research Institute, P.O. Box 358, DK-4000 Roskilde, Denmark;Department of Marine Ecology, Aarhus University, Finlandsgade 14, DK-8200 Aarhus, Denmark
Jesper H. Andersen2 and Gunni ÆrtebjergDepartment of Marine Ecology, National Environmental Research Institute, P.O. Box 358, DK-4000 Roskilde, Denmark
Abstract
In the past 2 decades significant measures have been taken to reduce nitrogen and phosphorus discharges fromDenmark by 50% and 80%, respectively. These nutrient reduction targets now appear within reach after severalconsecutive reduction measures are fully implemented, particularly toward diffuse discharges, and reduced nutrientconcentrations are beginning to be observed in estuaries and the Danish straits. Phosphorus concentrations havedeclined by 22% to 57% from the early 1990s, mainly owing to improved treatment of urban and industrialwastewater. Changes in nitrogen concentrations, following reduction measures toward diffuse sources, were morerecent and partly masked by large interannual variations in freshwater discharge. The response in marine nitrogenconcentrations was delayed relative to the decline in riverine concentrations, most likely owing to large internalloading from the sediments. Two consecutive dry years appeared to be the triggering mechanism for nitrogenconcentrations to decline. In the last 5 yr, nitrogen levels in estuaries and coastal waters have decreased up to 44%when interannual variations in freshwater discharge were accounted for. These first signs of environmental recoverywere most pronounced in estuaries and coastal waters but also were apparent in open waters of the Kattegat, theSound, and the Belt Sea. This case study is the first to document significant decreases in nutrient concentrations ona large regional scale resulting from an active management strategy to reduce nutrients from both diffuse and pointsources.
Eutrophication of coastal ecosystems from nutrient over-enrichment is widespread (Nixon 1995), with the effectsmanifested in a myriad of direct and indirect responses(Cloern 2001). Although the sources and pathways of nutri-ent inputs to aquatic ecosystems can be estimated with rea-sonable certainty, it has been difficult to achieve reductionsin the different sources (Boesch 2002). However, somecoastal ecosystems have experienced reductions in inputs ofphosphorus and nitrogen primarily through improvement in
1 Corresponding author ([email protected]).2 Present address: DHI Water & Environment, Agern Alle 5, DK-
2970 Hørsholm, Denmark.
AcknowledgmentsThe present work is a contribution of the CHARM (EVK3-CT-
2001-00065) and REBECCA (SSPI-CT-2003-502158) projectsfunded by the European Commission. We gratefully acknowledgethe Danish counties responsible for data collection under the DanishNationwide Aquatic Monitoring and Assessment Program and theSwedish Hydrological and Meteorological Institute for providingdata from the Swedish monitoring programs. German loading datawere provided by Heike Herata, Federal Environmental Agency inBerlin, and Thorkild Petenati from the federal state of Schleswig-Holstein. We thank Ole Hertel for providing atmospheric depositiondata and Bo Riemann for comments on the manuscript. The man-uscript was improved by valuable comments made by three anon-ymous reviewers.
treatment of wastewater and reductions in point sources(Butt and Brown 2000; Conley et al. 2000), although rela-tively little progress has been made in reducing diffusesources of nutrients (Butt and Brown 2000; Boesch 2002).In eastern Europe a number of studies have shown decreas-ing nutrient concentrations in rivers and streams from re-duced application of fertilizers after the economic collapsein eastern Europe and the Soviet Union in the early 1990s(for overview, see Stalnacke et al. 2003).
In 1987 a National Action Plan on the Aquatic Environ-ment was enacted in Denmark to reduce nutrient inputs tothe aquatic environment. This action plan was based on anagenda adopted in 1986 that aimed to reduce nitrogen andphosphorus discharges by 50% and 80%, respectively (Kron-vang et al. 1993). The commitment to nutrient reductionswas also made in multijurisdictional agreements with bothHelsinki Commission for the Protection of the Marine En-vironment of the Baltic Sea Area (HELCOM) and Oslo-ParisCommission for the Protection of the Marine Environmentof the North-East Atlantic (OSPAR) (Conley et al. 2002b).This first action plan was most effective toward nutrient re-ductions from municipal wastewater, and it was soon rec-ognized that new actions had to be taken toward the diffuseloading, in particular nitrogen. Another action plan for sus-tainable agricultural production followed in 1991 and a sec-ond National Action Plan on the Aquatic Environment in
399Eutrophication trend reversal
Fig. 1. Map of the Kattegat, the Sound, and the Belt Sea show-ing location of monitoring stations used in the study, partitionedinto estuarine and coastal stations (squares) and open-water stations(circles). Boundaries of the study area are marked by dashed lines.
Table 1. Catchment area and land use for the Kattegat, the Sound, and the Belt Sea. Areas are from HELCOM (2002), and land useswere compiled from the GRID-Arendal database (Sweitzer et al. 1996).
Denmark
Catchmentarea (km2)
Arable andpasture (%)
Sweden
Catchmentarea (km2)
Arable andpasture (%)
Germany
Catchmentarea (km2)
Arable andpasture (%)
KattegatThe SoundBelt Sea
15,8501,740
12,340
634768
20,920*2,885—
1864—
——
10,400†
——65
Total 29,930 64 23,805 23 10,400 65
* Excluding the catchment of the Gota River (50,233 km2).† The catchment area for the German federal state Schleswig-Holstein is 5,450 km2.
1998. The two latter action plans included a variety of strat-egies and measures to reduce diffuse nitrogen inputs, in-cluding fertilizer reductions, buffer strips, and restoration ofwetlands (for further details, see Conley et al. 2002b).
The Danish National Aquatic Monitoring and AssessmentProgram (DNAMAP) was established in 1988 to monitornutrient loading and ecological responses to the nutrient re-duction targets. DNAMAP was organized with the aim ofobtaining information on a wide range of eutrophication-related variables (e.g., nutrients, chlorophyll a, macrophytes,benthic macrofauna) covering many estuaries and coastalzones in Denmark. Monitoring in the open waters and se-lected coastal waters is a requirement of HELCOM and OS-PAR, with DNAMAP and the Swedish national monitoringprogram operating in a coordinated joint effort.
The objective of this article is to demonstrate that the
measures that have been taken in Denmark to reduce theload of nutrients from both point and nonpoint sources aresuccessful. We have analyzed the trends in nutrient loadingand concentrations after the first Action Plan on the AquaticEnvironment and the establishment of DNAMAP in 1989 upto the most recent data from 2002, a period of 14 yr withlarge changes in nutrient loading that has allowed us to iden-tify responses in the ecosystem to nutrient management. Thisanalysis was possible owing to the extensive data set col-lected under DNAMAP, and our report is the first documen-tation of significant effects in the marine ecosystem that canbe traced to nutrient reductions resulting from an active man-agement strategy.
Study area—The Kattegat, the Sound, and the Belt Sea(the Danish straits) comprise a shallow transition area be-tween the brackish Baltic Sea and the more saline Skagerrak/North Sea (Fig. 1). It is a coastal ecosystem with estuarinecharacter dominated by advective transports and an almostpermanent halocline located at ;15-m depth (Andersson andRydberg 1993; Jakobsen and Trebuchet 2000). Transport inthe surface layer is generally northward, whereas Skagerrakwater penetrates southward into the Danish straits as a bot-tom current that gradually mixes with the surface layer. Theresidual flow from the Baltic Sea is ;500 km3 yr21 (Stige-brandt and Gustafsson 2003), and bottom water exchangeswith the Baltic Sea occur over the sills at Drogden (8-mdepth) and Darss (15-m depth). The area and volume of theDanish straits are 41,000 km2 and 810 km3, respectively(Gustafsson 2000).
A total land area of 64,135 km2 discharges directly intothe Danish straits with 47%, 37%, and 16% belonging toDenmark, Sweden, and Germany, respectively (Table 1). TheGota River (Fig. 1), which is the sixth largest river in theentire Baltic Sea area, is not included in the figures for theKattegat, because discharges from this river mainly are car-ried northward out of the Kattegat into the Skagerrak. TheDanish straits receive discharges from 70% of Denmark. Ap-proximately 9 million people inhabit the catchment, the ma-jority of these living in urban settlements. Land use is gen-erally dominated by agriculture (Table 1), except for theSwedish catchment discharging into the Kattegat with 72%forest.
Local inputs of freshwater and nutrients are primarily dis-charged through productive estuaries and coastal regions(Carstensen et al. 2003). The Danish estuaries are for the
400 Carstensen et al.
most part shallow (,3 m deep) with relatively short resi-dence times (Conley et al. 2000). The majority of estuarieshave a well-mixed water column with intermittent periodsof stratification during periods of calm winds or inflow ofsaline bottom water. Agricultural production in Denmark isvery specialized and highly and consistently productive bothper unit land and per unit resource (Porter and Petersen1997). Over the past decade, there has been an increase inanimal husbandry, which together with the measures in theaction plans has precipitated a shift from chemical fertilizersto manure for crop production. Denmark is now the world’slargest exporter of pork meat, with a standing stock of 13million pigs in addition to 1.7 million cattle (2003 data fromwww.ddl.dk). For comparison, the human population ofDenmark is 5.4 million.
Materials and methods
Detailed load compilations have been carried out in Den-mark since 1989 as part of the first Action Plan for theAquatic Environment. Data on the freshwater dischargesfrom the Swedish catchment were obtained from the Swed-ish Meteorological and Hydrological Institute (SMHI), andnutrient loading figures were compiled from Stalnacke et al.(1999) and data from the Swedish University of AgriculturalSciences (www.slu.se). A long time series was available fornutrient and freshwater inputs from the German federal stateof Schleswig-Holstein (1977–2002), whereas total inputsfrom Germany to the Belt Sea were available from 1994onward (data source, Federal Environmental Agency, Berlin,Germany). We calculated the average ratio between totalGerman input and that from Schleswig-Holstein for 1994–2002 and used this value for scaling up the inputs fromSchleswig-Holstein during 1989–1993. Danish nutrient load-ing was partitioned into riverine and point source contribu-tions, the latter combined of discharges directly to marinewaters and discharges to freshwater carried with the riverineinput. The diffuse nutrient loading was calculated as riverineinput minus the point source input to freshwater. In this cal-culation we assumed that the retention of nutrients frompoint sources to freshwater was negligible, because fresh-water point sources generally discharge in the downstreamarea. It should be recognized that certainty in the loadingcompilations is likely to have increased with time. The ratiobetween the diffuse nutrient loading and the freshwater dis-charge will hereafter be referred to as flow-weighted con-centrations of TN and TP.
Atmospheric nitrogen deposition was calculated by meansof a Lagrangian model with 96-h trajectories of air parcelsto a net of receptor points having a resolution of 25 3 25km (Hertel et al. 1995). The model was calibrated to depo-sition rates measured at two coastal gauges located in thenorthern and southern part of the study area. Atmosphericdeposition of phosphorus has not been calculated on an an-nual basis, but Andersen et al. (2004) estimated it to be ;8kg P km22 for the study area and contended that temporaltrends were unlikely. Based on these results a constant at-mospheric phosphorus input of 328 3 103 kg yr21 was as-sumed in this study.
Measurements of nutrient concentrations (NH , NO ,1 24 2
NO , PO , TN, and TP), collected within the framework2 323 4
of DNAMAP and the Swedish national monitoring program,were investigated in the present study. A total of 46 stationslocated in estuaries and the coastal region and 27 open-waterstations (Fig. 1) were sampled with varying frequencies fromone up to 103 times per year, however unevenly distributedboth within and between years. Dissolved inorganic nitrogen(DIN) was calculated as the sum of ammonia, nitrite, andnitrate, whereas dissolved inorganic phosphorus (DIP) com-prised phosphate only. For estuarine and coastal stations, wecalculated the average concentration of all nutrient constit-uents over the entire water column, whereas average con-centrations of samples #10 m and samples $20 m wereused to characterize the surface and the bottom layer at open-water stations.
Yearly means of DIN, DIP, TN, and TP were computedthrough use of a general linear model that described varia-tions between stations, years, and months after log-transfor-mation of the variables. Thus, if Yijkl described the obser-vations of any of the four considered variables, then
Y 5 station 3 year 3 month 3 «ijkl i j k ijkl
⇑⇓log(Y ) 5 log(station ) 1 log(year ) 1 log(month )ijkl i j k
1 e (1)ijkl
where eijkl was a normal distributed random error with zeromean and variance s2, stationi described the station-specificmean levels, yearj described the year-specific mean level,and monthk described the seasonal variation by monthlymeans. It was assumed that interannual and seasonal varia-tions were multiplicative factors to each other and to thestation-specific mean level, and that the error term on theoriginal scale was lognormaly distributed.
The monitoring data were not balanced, i.e., uneven num-ber of observations for different combinations of station,year, and month, and averaging all observations for a givenyear would result in values that depended on the differencesin sampling frequencies. Comparable yearly means were cal-culated by computing the marginal distributions of yearj aslinear combinations of the parameter estimates in Eq. 1(Searle et al. 1980) to account for differences in samplingfrequencies. Yearly means obtained from Eq. 1 were back-transformed to the original scale by
2sE(year ) 5 exp E[log(year )] 1j j5 62
In the following we shall refer to the back-transform of themarginal means computed from the model in Eq. 1 as theyearly nutrient means or levels.
Nitrogen and phosphorus concentrations were related tofreshwater discharge, and point source nitrogen and phos-phorus loading, respectively, by means of multiple linear re-gression models using yearly means from 1989–1997. Thelast 5 yr (1998–2002) were used to investigate deviationsfrom those established relationships that could potentiallyaccrue from changes in agricultural practices. Mean nutrientconcentrations based on estuarine and coastal stations were
401Eutrophication trend reversal
Fig. 2. Annual discharges of (A) freshwater, (B) total nitrogen,and (C) total phosphorus to the Danish straits (1989–2002). At-mospheric nitrogen and phosphorus depositions to the study areaare shown by diamonds connected with a thin line. Danish nitrogenand phosphorus input has been partitioned into point sources (belowdashed line) and diffuse sources (above dashed line). Inputs fromthe Gota River were not included in the Swedish figures.
Table 2. Trend analysis of nutrient inputs to the Danish straits (1989–2002, n 5 14) by linear regression (F1,12). Significant trends(103 kg yr21) at the 5% significance level are highlighted by boldface type.
Source
Denmark
Trend p
Sweden
Trend p
Germany
Trend p
TN diffuse sourcesTN point sourcesTN total input
3542153721184
0.7558,0.0001
0.3088
——
212
——
0.9718
——193
——
0.5919TP diffuse sourcesTP point sourcesTP total input
2722812253
0.3933,0.0001
0.0001
——
216
——
0.0761
——
2
——
0.7048
related to input from Denmark only. In contrast, a combinedfreshwater discharge from Denmark, Sweden, and Germanywas used to relate to concentrations found at open-waterstations. We used point source loading from Denmark alonefor the open-water stations, because annual figures of nutri-ent loading partitioned into point and diffuse sources werenot available from Sweden and Germany.
Finally, the mean nutrient levels were adjusted to varia-tions in freshwater discharge and in point source loading(denoted DINADJ, TNADJ, DIPADJ, and TPADJ) by means of themultiple regressions models described above. Differencesbetween the yearly means and predicted values from themultiple regressions were added to the predicted mean levelsfor DIN, TN, DIP, and TP corresponding to the averagefreshwater discharge over the entire period (1989–2002) andthe average point source loading for the five most recentyears (1998–2002). The adjusted nutrient levels would in-dicate changes in the diffuse inputs when random variationsin freshwater discharge were taken into account. These ad-justed means described the nutrient level in a given year ifthe loading from point sources had been low and if the fresh-water discharge had an average level. Nutrient inputs, nutri-ent levels, and adjusted nutrient levels were analyzed fortrends by means of linear regression.
Results
Nutrient loading—There were strong interannual varia-tions in freshwater discharge as well as nutrient loading (Fig.2). The total freshwater discharge varied from 12–29 km3
yr21, and interannual variations in the Danish, Swedish, andGerman values were highly correlated (rDK,SE 5 0.82; rDK,GE
5 0.86; rSE,GE 5 0.64). Particularly, 1989, 1996, and 1997were ‘‘dry’’ years, whereas the other years had freshwaterdischarges .20 km3 yr21. During the entire period the fresh-water discharge from Sweden was the largest (49%), fol-lowed by discharge from Denmark (40%) and Germany(11%). However, the loading from Denmark was clearly thelargest for both total nitrogen (62%) and total phosphorus(74%). Interannual variation in the Danish, Swedish, andGerman contributions were highly correlated for nitrogenloading (rDK,SE 5 0.87; rDK,GE 5 0.79; rSE,GE 5 0.86) and lesscorrelated for phosphorus loading (rDK,SE 5 0.77; rDK,GE 50.22; rSE,GE 5 0.68). All freshwater and nutrient dischargesfrom the three countries were significantly correlated (p ,0.05), except rDK,GE for phosphorus loading (p 5 0.4552).
Interannual variations in total nitrogen loading from landdid not reflect any trends (Table 2) and were clearly linkedto freshwater discharge (Fig. 2). Over the entire study period,the input from Danish point sources declined significantly,comprising ;50% in the dry year of 1989 to ,10% of thetotal Danish contribution in the most recent years. This cor-responded to a reduction of ;20,000 3 103 kg of nitrogen.The total phosphorus loading decreased significantly from
402 Carstensen et al.
Fig. 3. The ratio between the diffuse nutrient loading and thefreshwater discharge from Denmark, referred to as flow-weightedTN and TP concentrations in the text.
;6000 3 103 kg yr21 in the beginning of the period to;3000 3 103 kg yr21 after 1994. This trend was attributedto reductions in Danish point sources from ;4500 3 103 kgyr21 in 1989 to 600–800 3 103 kg yr21 in recent years. Pointsource reductions have made the diffuse input of phosphorusthe dominating source in Denmark from 41% in 1989 to;80% in recent years. This change in dominating inputsfrom point to diffuse sources has also resulted in a gradualcovariation of phosphorus loading with the freshwater dis-charge. There were no significant correlations between thefreshwater discharges and point source loading data used inthe multiple regression analysis below (t-test, all p . 0.05).
Atmospheric deposition of nitrogen was relatively stable(average of 55,000 3 103 kg yr21) over the entire study pe-riod, ranging from 47,000 3 103 kg yr21 to 65,000 3 103
kg yr21 with actually only 1 yr (1990) exceeding 60,000 3103 kg yr21 (Fig. 2). There was no trend in atmospheric de-position (F1,12 5 3.66, p 5 0.0799), particularly if 1990 wasexcluded in the linear regression (F1,11 5 1.55, p 5 0.2394).Consequently, atmospheric nitrogen deposition was consid-ered constant over the study period and not included in themultiple regression. Atmospheric deposition of phosphoruswas on average ,10% of the land-based inputs and was notincluded in the multiple regression for the same reason.
The ratio between nitrogen and phosphorus diffuse inputand freshwater discharge had the largest variations in thebeginning of the period when loading compilations wereconsidered more uncertain (Fig. 3). There were no significanttrends in the flow-weighted concentrations for either TN(F1,12 5 2.28, p 5 0.1573) or TP (F1,12 5 0.06, p 5 0.8136)over the entire period, but TN decreased significantly (F1,11
5 10.90, p 5 0.0071) if the first year with more uncertainloading figures was excluded. Flow-weighted TN concentra-tions from diffuse sources was about 7 mg L21 in the begin-ning of the 1990s, decreasing to ;5.5 mg L21 in recentyears. Low levels were observed in the three dry years of1989, 1996, and 1997. The flow-weighted TP concentrationfrom diffuse sources was relatively constant, ;0.11 mg L21
from 1993 and onward. There was no trend in TP levelsduring this period (F1,8 5 0.95, p 5 0.3594).
Nutrient concentrations—The two wet years in 1994 and1995 and the two dry years in 1996 and 1997 were clearlyvisible in the mean nitrogen levels in estuaries and coastalareas as well as for the open-water stations (Fig. 4A,C). DIN
levels in surface waters decreased by ;30% from the twofirst years to the two last years, whereas there was no changein bottom water concentrations. TN levels decreased by 12–18% during the study period. However, nitrogen means dur-ing the last couple of years were almost at the same levelas in 1996 and 1997 although the freshwater discharge wasconsiderably higher. Trends were not significant for DIN,whereas TN levels decreased significantly by 8 mg L21 yr21
in estuaries and coastal waters and approximately at half therate in open waters (Table 3).
The effect of dry and wet years on phosphorus levels wasless pronounced; the phosphorus means decreased in the be-ginning of the study period and were more or less stationaryin the last part of the investigated period (Fig. 4B,D). DIPlevels decreased with significant changes in surface watersof 48–57% observed between 1989–1990 to 2001–2002 (Ta-ble 3), whereas the decline in the open-water bottom layerwas more moderate (22%). Trends in TP levels were alsosignificant with more similar changes using the same periods(30–39%) for the three considered water types.
The yearly means of DIN and TN were significantly re-lated to freshwater discharge for estuaries and coastal sta-tions as well as for the surface and bottom layer at open-water stations (Table 4). Nitrogen loading from point sourcesdid not show a consistent pattern for explaining interannualvariations in nitrogen concentrations, with significant rela-tionships observed only for DIN levels in the surface layerof open-water stations and TN levels in the bottom layer ofopen-water stations. The yearly means of DIP and TP wereclearly linked to point source loadings of phosphorus (Table4), whereas interannual variations in freshwater dischargesdid not correlate significantly. The most significant relation-ships were obtained for estuarine and coastal stations, allhaving R2-values .0.85.
In the last 5 yr of the study period, nitrogen did not exhibitthe same behavior with respect to freshwater discharge asfor the earlier period of 1989–1997 (Fig. 5). All the yearlymeans for both DIN and TN were below the regression linesexcept for the DIN level in the bottom layer at open-waterstations that had 2 yr above the regression line (1999 and2002). Although the phosphorus point source loading fromDenmark was lower in 1998–2002 than in all the previousyears, the phosphorus levels in these years were mostlyabove the extrapolation of the regression lines, particularlyfor DIP in estuaries and the coastal area as well as the open-water surface layer (Fig. 6).
Nitrogen levels adjusted for variations in freshwater dis-charge and nitrogen point source loading showed consistentdecreasing trends (Fig. 7A,C), all significant but the adjustedDIN means at open-water stations (Table 3). Removing theinterannual variation related to freshwater discharge im-proved the significance of the trends. The relative change inadjusted DIN levels from 1989–1990 to 2001–2002 variedfrom 23% in the open-water bottom layer to 14% in theopen-water surface layer and 44% in estuaries and coastalwaters. These trends were more similar for adjusted TN lev-els (15–18%). Adjusted phosphorus levels were generallymore variable than were adjusted nitrogen levels (Fig.7B,D), however, without any significant trend (Table 3). Forboth nitrogen and phosphorus, the highest rate of change and
403Eutrophication trend reversal
Fig. 4. Yearly means for estuarine and coastal stations, surface layer of open-water stations, and bottom layer of open-water stations.Error bars show the 95% confidence limits of the mean values.
Table 3. Trend analysis of the nutrient means (1989–2002, n 5 14) and means adjusted for variations in freshwater discharge and pointsource loading. Significant trends (mg L21 yr21) at the 5% significance level are highlighted by boldface type.
Variable
Estuarine and coastal
Trend p
Open-water surface
Trend p
Open-water bottom
Trend p
DINTNDIPTP
23.9328.1721.7822.05
0.10840.0321
,0.0001,0.0001
20.6022.9420.4220.87
0.07230.01430.0040
,0.0001
20.2224.5520.7021.15
0.86510.01340.0032
,0.0001DINADJ
TNADJ
DIPADJ
TPADJ
26.36211.02
0.160.22
0.00060.00030.40230.2783
20.1323.70
0.1120.17
0.48990.00050.27530.1970
20.3424.04
0.2320.08
0.68510.00050.14350.5514
the most significant trends were observed in estuaries andcoastal regions.
Discussion
In this study we have identified strong relationships be-tween land-based discharges and nutrient concentrations inthe marine environment. This was possible owing to severalreasons. First, yearly means of nutrient concentrations witha high precision were obtained by pooling data from a large
number of stations, assuming that all stations had the esti-mated interannual variation in common. Second, the inves-tigated period had large variations in both freshwater dis-charges and point source phosphorus loadings, yielding ahigh power for the multiple regression analysis. Third, in-terannual variations in freshwater discharge and point sourceloading were not correlated, and the estimates resulting fromthe multiple regression were consequently not biased. Final-ly, changes in nutrient concentrations attributed to manage-ment actions occurred at different times for nitrogen (end of
404 Carstensen et al.
Table 4. Nutrient means (1989–1997, n 5 9) related to freshwater discharge and nutrient loading from point sources by multipleregression. Freshwater discharge included only Danish data for estuarine and coastal stations, and the contribution from Denmark, Sweden,and Germany for open-water stations. Total nitrogen loading from Denmark was used for DIN and TN levels, and total phosphorus loadingfrom Denmark was used for DIP and TP levels. Significant relations (F1,6 at 5% significance level) are highlighted by boldface type.
Variable R2 Intercept
Freshwater discharge
Estimate(mg L21 km23) p
Point source loading
Estimate(mg L21 1026 kg21) p
DINEstuarine and coastal stationsOpen-water stations (surface)Open-water stations (bottom)
0.95350.75670.6610
33.86477
42.74
16.830.6042.615
,0.00010.03200.0211
0.8050.4390.510
0.17070.03550.4491
TNEstuarine and coastal stationsOpen-water stations (surface)Open-water stations (bottom)
0.95090.87840.9243
416.9237.5190.8
25.5563.1735.039
,0.00010.00070.0003
1.7630.2141.463
0.07500.59000.0289
DIPEstuarine and coastal stationsOpen-water stations (surface)Open-water stations (bottom)
0.87010.69770.7676
12.884.59
18.13
20.204720.035020.0735
0.72970.57950.5294
6.771.843.25
0.00070.01000.0044
TPEstuarine and coastal stationsOpen-water stations (surface)Open-water stations (bottom)
0.92280.76370.8650
34.3416.4028.62
0.30190.12220.0313
0.58440.44970.8424
8.322.633.86
0.00020.00610.0009
Fig. 5. Yearly (A) DIN and (B) TN levels for estuarine andcoastal stations, surface layer of open-water stations, and bottomlayer of open-water stations versus freshwater discharge. Freshwaterdischarges related to nitrogen levels for open-water stations includ-ed contributions from Denmark, Sweden, and Germany; coastal ni-trogen levels were related to Danish discharges only. Open symbolsshow the recent levels (1998–2002) not included in the multipleregression.
Fig. 6. Yearly (A) DIP and (B) TP levels for estuarine andcoastal stations, surface layer of open-water stations, and bottomlayer of open-water stations versus phosphorus point source loadingfrom Denmark. Open symbols show the recent levels (1998–2002)not included in the multiple regression.
405Eutrophication trend reversal
Fig. 7. Nutrient means adjusted for variations in both freshwater discharge and point source loading from the regression analysis.
the 1990s) and phosphorus (beginning of the 1990s), allow-ing us to separate out these different sources of variation(Conley et al. 2000)
Trends in nutrient concentrations—Nitrogen concentra-tions showed decreasing trends (Table 3) when variations infreshwater discharge were accounted for, and this decline canprimarily be attributed to changes in diffuse loading. This isalso shown by decreasing flow-weighted TN concentrations(Fig. 3), taking into account that nitrogen loading in the be-ginning of the period was more uncertain and that the highvalue in 1998 may be a consequence of the two dry yearsof 1996 and 1997 with nitrogen accumulating in the catch-ment.
Although flow-weighted concentrations of TN began todecline already in 1994 (Fig. 3), the response in nitrogenlevels was not clearly identifiable until 1998 in the meanlevels adjusted for variations in freshwater discharge (Fig.7A,C). This delayed response may be partly owing to a sub-stantial reduction in the internal recycling of nitrogen fol-lowing the two dry years. In 1994 and 1995, nitrogen levelsin the water column were, most likely, kept high by a largeinternal nitrogen release from the sediments. This may po-tentially also have been the case in 1996 and 1997, althoughthese two dry years constitute one end of the scale in therelationship between nitrogen concentrations and freshwater
discharge (Fig. 5). Coupling between nutrient loading, water-column production of organic matter, and recycling of nu-trients from sediments occurs over time scales of severalyears or less (Boynton et al. 1995). Attempts to find signif-icant correlations between nutrient load and system level re-sponses in estuaries often succeeds only when annual nutri-ent loads are combined with some fraction of the previousyear’s nutrient load (Boynton et al. 1995; Conley et al.2000), suggesting that an internal load is important.
Phosphorus concentrations declined substantially in estu-aries and coastal areas as well as in the open-waters of theDanish straits during the beginning of the investigated period(Fig. 4B,D). This trend was clearly linked to reductions inpoint source loading, mainly through improved wastewatertreatment. Changes in diffuse phosphorus loading should inprinciple, as shown for point source loading, show a similarvariation in phosphorus concentrations in the water column,but these small-scale variations related to freshwater dis-charge are masked by larger fluxes such as phosphorus re-lease during anoxic conditions from sediments. In the BalticSea, for example, annual variations in phosphorus releasefrom sediments with variations in anoxia are over an orderof magnitude larger than annual phosphorus loading (Conleyet al. 2002a).
Flow-weighted TP concentrations in freshwater dischargecorrected for point sources have remained almost constant,
406 Carstensen et al.
Fig. 8. The nitrogen/phosphorus molar ratio for the nutrient means (A) and nutrient means adjusted to variations in freshwater discharge(B).
whereas flow-weighted TN concentrations have declined.This has naturally altered the nitrogen/phosphorus ratio ofdiffuse loading and consequently that of the total loading.The increase in the DIN/DIP ratio in the beginning of the1990s (Fig. 8) signals a change from potential nitrogen lim-itation in favor of phosphorus limitation. In the last 5 yr, theDIN/DIP ratio has declined in estuaries and coastal waters,which may potentially have led to more nitrogen limitationin favor of phosphorus limitation. Several Danish estuarieshave shown spring phosphorus limitation switching to nitro-gen limitation in early summer (Holmboe et al. 1999). Therecan still be large interannual variations in nitrogen and phos-phorus limitation owing to changes in freshwater discharge,but nitrogen has become more important as the limiting nu-trient over the last 5 yr (Fig. 8B). Although the decreasingtrends in DIN and DIP may have resulted in changing pat-terns of nutrient limitation, it should be acknowledged thatthe combined effect has increased overall nutrient limitation(Ærtebjerg et al. 2003).
Bottom water concentrations—Advective transport dom-inates the open-water of the Danish straits, and the origin ofthe inflowing water from bottom waters in the Skagerrakmay originate from different regions of the North Sea, withlarge variations in nutrient levels (Rydberg et al. 1996). Themain source of inflowing bottom water is from the centralNorth Sea having moderate nutrient levels and salinity of;34, but occasionally nutrient-rich water with salinity of;32 originating from the German Bight and carried with theJutland Coastal Current spills into the Kattegat (Rydberg etal. 1996). This has happened to varying degrees in 1989,1995, 1999, and 2002 (observed in winter–spring monitoringdata of DIN vs. salinity) (data not shown), which could ex-plain the relatively high values of adjusted DIN levels inthese years (Fig. 7A). Another phenomenon that may influ-ence bottom water concentrations in the Danish straits, par-ticularly in the Kattegat, is the inflow of DIN-depleted sur-face water from the Skagerrak into the bottom layer(observed in 1990 and 1997) (data not shown).
The phosphorus pool in the bottom layer is considered todepend on local loading, water exchanges, and oxygen con-
ditions (Rasmussen et al. 2003b). Oxygen depletion is a re-occurring phenomenon in the Danish straits (Andersson andRydberg 1988, 1993; Babenerd 1991), and 2002 was theworst year ever recorded, with 21% of the area having ox-ygen concentrations ,2 mg L21 for extensive periods (HEL-COM 2003). On the other hand, oxygen conditions weregenerally good in 1997 (Rasmussen et al. 2003a). These 2yr, corresponding to the highest and lowest values for theadjusted DIP means, demonstrate that sediment phosphorusrelease during anoxic conditions increases the DIP levels inthe bottom layer in open waters (Mortimer 1941; Conley etal. 2002a).
Nutrient levels in the open-water bottom layer were main-ly determined by local inputs of both nitrogen and phospho-rus, with relationships between concentrations and loads asstrong as those in the open-water surface layer (Table 3).These interannual variations in nutrient levels cannot be ex-plained by inflow from the central part of the North Sea,where nutrient levels are low (OSPAR Commission 2000)and presumably not directly influenced by land-based load-ing. This suggests a substantial vertical exchange of nutrientsover the pycnocline through upwelling and entrainment(Gustafsson 2000) and remineralization of sedimenting par-ticulate matter, particularly after the diatom spring bloom(Josefson and Hansen 2003). This supports the idea of theDanish straits being a marginal sea with estuarine character.Thus, interannual variations in bottom water nutrient levelswere mainly determined by discharges from local sources,whereas changes in Skagerrak inflow and episodes of oxy-gen depletion only caused minor deviations from this pat-tern.
Nutrient management—Over the past 2 decades, coastaleutrophication of Danish marine waters has been a majorconcern, and substantial nutrient reductions have beenachieved through national action plans, international marineconventions, and European Union legislation (Iversen et al.1998; Conley et al. 2000; Ærtebjerg et al. 2003). The de-clining trends in nutrient concentrations documented here areto our knowledge the first successful large-scale effort toreduce inputs from both point and diffuse sources.
407Eutrophication trend reversal
Table 5. Nitrogen and phosphorus inputs to the aquatic environment for the baseline (mid-1980s) with reduction targets partitioned intosectors. UWTPs, urban wastewater treatment plants; IDs, industrial discharges. Phosphorus discharges from the agricultural sector includelosses from farmyards only. For details, please see Ærtebjerg et al. (2003) and Grant and Waagepetersen (2003).
Sector
Nitrogen
Baseline 4 Reduction (%) 5 Target
Phosphorus
Baseline 4 Reduction (%) 5 Target
AgricultureUWTPsIDs
311,000 4 152,400 (49) 5 158,60018,000 4 11,400 (60) 5 6,600
5,000 4 3,000 (60) 5 2,000
4,400 4 4,000 (91) 5 4004,470 4 3,250 (72) 5 1,2201,250 4 1,050 (82) 5 1,820
Total 334,000 4 166,800 (50) 5 167,200 10,120 4 8,300 (80) 5 3,440
The first national initiative was the 1985 NPo Action Planwith a suite of measures implemented in relation to the dis-charge of nitrogen (N), phosphorus (P), and organic matter(o) from agriculture and wastewater; however, specific re-duction targets were not set. An event of widespread hypoxiain the Danish straits in 1986 led to the adoption of an agendaurging the government to reduce discharges and losses ofnitrogen (by 50%) and phosphorus (by 80%) from agricul-ture, municipal wastewater treatment plants, and individualindustrial discharges (Kronvang et al. 1993; Conley et al.2002b). This strategic aim was formulated into sector-spe-cific reduction objectives and targets for (1) discharges andlosses from agriculture, (2) discharges from municipalwastewater treatment plants, and (3) direct discharges fromindustries (Table 5).
The reduction targets for both municipal wastewater treat-ment plants and industries were met in 1995, whereas thespecific objectives and targets for the agricultural sector weredifficult to meet within the original time frame. The ActionPlan on Sustainable Agriculture adopted in 1991, focusingon reduction of losses from cultivated fields, was followedby the second Action Plan on the Aquatic Environment withadditional measures in 1998 to fulfill the requirements of theEuropean Union Nitrates Directive (Anonymous 1991). Thedecision for the second action plan was influenced by col-lapse of the Mariager Fjord estuary, which went completelyanoxic in 1997 (Fallesen et al. 2000).
Total expected reductions in nitrogen root zone lossesfrom agriculture in 2002 were estimated at 149,000 3 103
kg, corresponding to a reduction of 48% in the most recentassessment on the effectiveness of the measures (Grant andWaagepetersen 2003). However, the reduction estimates areassociated with substantial uncertainty, and there are consid-erable time lags between changes in agriculture practice andwater quality responses (Stalnacke et al. 2003; Tomer andBurkart 2003). These reductions would not have beenachieved if the periodic assessment for reduction compliancehad not been carried out and if the national monitoring pro-gram had not maintained focus on eutrophication. Reduc-tions in nutrient overenrichment of coastal ecosystems willrely on implementation of an adaptive management frame-work (Boesch 2002). Further reductions in diffuse nitrogenloading may be needed, considering that Denmark has hadone of the highest area-specific nitrogen loss rates (Conleyet al. 2000) and that the Danish straits with an almost per-manent stratification are prone to hypoxia.
Ecosystem perspectives—Decreasing nutrient concentra-tions is the first step toward reducing the adverse effects ofeutrophication. Danish estuaries and coastal areas have hada long record of eutrophication symptoms from nutrient en-richment, including increased primary production (Richard-son and Heilmann 1995), decreasing bottom oxygen con-centrations (Andersson 1996), and loss of macrophytes(Borum and Sand-Jensen 1991). Interannual variations in ni-trogen loading are reflected in summer chlorophyll a con-centration and bloom frequency in the Kattegat (Carstensenet al. 2004) as well as primary production (Carstensen et al.2003). Nielsen et al. (2002b) found that chlorophyll a andwater transparency were significantly related to total nitro-gen concentrations in Danish estuaries and coastal watersand, further, that the depth colonization of eelgrass and ma-croalgae were significantly related to water transparency(Nielsen et al. 2002a). These small components of the largecomplex ecosystem show that reduced nutrient concentra-tions are likely to improve the ecological status of estuariesand coastal and open waters in Denmark through direct ordelayed responses or alternatively through threshold mech-anisms (Scheffer et al. 2001). Some signs of ecosystem re-covery have already been observed (Ærtebjerg et al. 2003).
References
ÆRTEBJERG, G., J. H. ANDERSEN, AND O. S. HANSEN [EDS.]. 2003.Nutrients and eutrophication in Danish marine waters: A chal-lenge for science and management. National EnvironmentalResearch Institute.
ANDERSEN, J. M., AND OTHERS. 2004. Aquatic environment 2003:State and trends—technical summary, technical report no. 500.National Environmental Research Institute.
ANDERSSON, L. 1996. Trends in nutrient and oxygen concentrationson the Skagerrak-Kattegat. J. Sea Res. 35: 63–71.
, AND L. RYDBERG. 1988. Trends in nutrient and oxygenconditions within the Kattegat: Effects of local nutrient supply.Estuar. Coast. Shelf Sci. 26: 559–579.
AND . 1993. Exchange of water and nutrients be-tween the Skagerrak and the Kattegat. Estuar. Coast. Shelf Sci.36: 159–181.
[ANONYMOUS]. 1991. Council directive 91/676/EEC of 12 Decem-ber 1991 concerning the protection of waters against pollutioncaused by nitrates from agricultural sources. Official Journal L375. European Commission.
BABENERD, B. 1991. Increasing oxygen deficiency in Kiel Bay(Western Baltic): A paradigm of progressing coastal eutrophi-cation. Meeresforschung 33: 121–140.
BOESCH, D. F. 2002. Challenges and opportunities for science in
408 Carstensen et al.
reducing nutrient overenrichment of coastal ecosystems. Estu-aries 25: 886–900.
BORUM, J., AND K. SAND-JENSEN. 1991. Interactions among phy-toplankton, periphyton, and macrophytes in temperate fresh-water and estuaries. Aquat. Bot. 41: 137–176.
BOYNTON, W. R., J. H. GARBER, R. SUMMERS, AND W. M. KEMP.1995. Inputs, transformations, and transport of nitrogen andphosphorus in Chesapeake Bay and selected tributaries. Estu-aries 18: 285–314.
BUTT, A. J., AND B. L. BROWN. 2000. The cost of nutrient reduction:A case study of Chesapeake Bay. Coast. Manage. 28: 175–185.
CARSTENSEN, J., D. J. CONLEY, AND P. HENRIKSEN. 2004. Frequen-cy, composition, and causes of summer phytoplankton bloomsin a shallow coastal ecosystem, the Kattegat. Limnol. Ocean-ogr. 49: 190–201.
, , AND B. MULLER-KARULIS. 2003. Spatial and tem-poral resolution of carbon fluxes in a shallow coastal ecosys-tem, the Kattegat. Mar. Ecol. Prog. Ser. 252: 35–50.
CLOERN, J. E. 2001. Our evolving conceptual model of the coastaleutrophication problem. Mar. Ecol. Prog. Ser. 210: 223–253.
CONLEY, D. J., C. HUMBORG, L. RAHM, O. P. SAVCHUK, AND F.WULFF. 2002a. Hypoxia in the Baltic Sea and basin-scalechanges in phosphorus biogeochemistry. Environ. Sci. Tech.36: 5315–5320.
, H. KAAS, F. MøHLENBERG, B. RASMUSSEN, AND J. WIN-DOLF. 2000. Characteristics of Danish estuaries. Estuaries 23:820–837.
, S. MARKAGER, J. ANDERSEN, T. ELLERMANN, AND L. M.SVENDSEN. 2002b. Coastal eutrophication and the Danish Na-tional Aquatic Monitoring and Assessment Program. Estuaries25: 706–719.
FALLESEN, G., F. ANDERSEN, AND B. LARSEN. 2000. Life, death andrevival of the hypertrophic Mariager Fjord, Denmark. J. Mar.Syst. 25: 313–321.
GRANT, R., AND J. WAAGEPETERSEN. 2003. Vandmiljøplan II: Slu-tevaluering [in Danish]. Report from National EnvironmentalResearch Institute and Danish Institute of Agricultural Scienc-es. Available from http://www.dmu.dk [Accessed June 2005]
GUSTAFSSON, B. 2000. Time-dependent modelling of the Baltic en-trance area, 1: Quantification of circulation and residence timesin the Kattegat and the straits of the Baltic sill. Estuaries 23:231–252.
HELCOM. 2002. Environment of the Baltic Sea area 1994–1998.Balt. Sea Environ. Proc. no. 82B.
. 2003. The 2002 oxygen depletion event in the Kattegat,Belt Sea and Western Baltic. Balt. Sea Environ. Proc. no. 90.
HERTEL, O., J. CHRISTENSEN, E. H. RUNGE, W. A. H. ASMAN, R.BERKOWICZ, M. F. HOVMAND, AND Ø. HOV. 1995. Develop-ment and testing of a new variable scale air pollution model—ADCEP. Atmos. Environ. 29: 1267–1290.
HOLMBOE, N., H. S. JENSEN, AND F. Ø. ANDERSEN. 1999. Nutrientaddition bioassays as indicators of nutrient limitation of phy-toplankton in an eutrophic estuary. Mar. Ecol. Prog. Ser. 186:95–104.
IVERSEN, T. M., R. GRANT, AND K. NIELSEN. 1998. Nitrogen en-richment of European inland and marine waters with specialattention to Danish policy measures. Environ. Poll. 102: 771–780.
JAKOBSEN, F., AND C. TREBUCHET. 2000. Observations of the trans-
port through the Belt Sea and an investigation of the momen-tum balance. Cont. Shelf Res. 20: 293–311.
JOSEFSON, A., AND J. L. S. HANSEN. 2003. Quantifying plant pig-ments and live diatoms in aphotic sediments of Scandinaviancoastal waters confirms a major route in the pelagic–benthiccoupling. Mar. Biol. 142: 649–658.
KRONVANG, B., G. ÆRTEBJERG, R. GRANT, P. KRISTENSEN, M. HO-VMAND, AND J. KIRKEGAARD. 1993. Nationwide monitoring ofnutrients and their ecological effects: State of the Danish aquat-ic environment. Ambio 22: 176–187.
MORTIMER, C. H. 1941. The exchange of substances between mudand water in lakes: I. J. Ecol. 29: 280–329.
NIELSEN, S. L., K. SAND-JENSEN, J. BORUM, AND O. GEERTZ-HAN-SEN. 2002a. Depth colonization of eelgrass (Zostera marina)and macroalgae as determined by water transparency in Danishcoastal waters. Estuaries 25: 1025–1032.
, , , AND . 2002b. Phytoplankton, nu-trients, and transparency in Danish coastal waters. Estuaries25: 930–937.
NIXON, S. W. 1995. Coastal marine eutrophication: A definition,social causes, and future concerns. Ophelia 41: 199–219.
OSPAR COMMISSION. 2000. Quality Status Report 2000, RegionII—Greater North Sea. OSPAR Commission.
PORTER, J. R., AND E. H. PETERSEN. 1997. Danish agriculture andits sustainability: A profile. Ambio 26: 462–465.
RASMUSSEN, B., B. G. GUSTAFSSON, G. ÆRTEBJERG, AND C. LUNDS-GAARD. 2003a. Oxygen concentration and consumption at theentrance to the Baltic Sea from 1975 to 2000. J. Mar. Syst. 42:13–30.
, , A. STOCKENBERG, AND G. ÆRTEBJERG. 2003b.Nutrient loads, advection and turnover at the entrance to theBaltic Sea. J. Mar. Syst. 39: 43–56.
RICHARDSON K., AND J. P. HEILMANN. 1995. Primary production inthe Kattegat past and present. Ophelia 41: 317–328.
RYDBERG, L., J. HAAMER, AND O. LIUNGMAN. 1996. Fluxes of waterand nutrients within and into the Skagerrak. J. Sea Res. 35:23–38.
SCHEFFER, M., S. CARPENTER, J. A. FOLEY, C. FOLKE, AND B.WALKER. 2001. Catastrophic shifts in ecosystems. Nature 413:591–596.
SEARLE, S. R., F. M. SPEED, AND G. A. MILLIKEN. 1980. Populationsmarginal means in the linear model: An alternative to leastsquares means. Am. Stat. 34: 216–221.
STALNACKE, P., A. GRIMVALL, C. LIBISELLER, M. LAZNIK, AND I.KOKORITE. 2003. Trends in nutrient concentrations in Latvianrivers and the response to the dramatic change in agriculture.J. Hydrol. 283: 184–205.
, , K. SUNDBLAD, AND A TONDERSKI. 1999. Estima-tion of riverine loads of nitrogen and phosphorus to the BalticSea, 1970–1993. Environ. Monit. Assess. 58: 170–200.
STIGEBRANDT, A., AND B. G. GUSTAFSSON. 2003. Response of theBaltic Sea to climate change: Theory and observations. J. SeaRes. 330: 1–14.
SWEITZER, J., S. LANGAAS, AND C. FOLKE. 1996. Land cover andpopulation density in the Baltic Sea drainage basin: A GISdatabase. Ambio 25: 191–198.
TOMER, M. D., AND M. R. BURKART. 2003. Long-term effects ofnitrogen fertilizer use on groundwater nitrate in two small wa-tersheds. J. Environ. Qual. 32: 2158–2171.
Received: 19 March 2004Accepted: 29 April 2005
Amended: 1 May 2005
EUTROPHICATION IN COASTAL ECOSYSTEMS
Eutrophication in coastal marine ecosystems: towards betterunderstanding and management strategies
J. H. Andersen Æ D. J. Conley
Published online: 27 April 2009
� The Author(s) 2009. This article is published with open access at Springerlink.com
The Second International Symposium on Research
and Management of Eutrophication in Coastal Eco-
systems took place 20–23 June 2006 in Nyborg,
Denmark. The Symposium was attended by more
than 200 persons with a specific interest in eutrophi-
cation processes as well as a common interest in
science-based management and implementation of
nutrient reduction strategies. More than 120 oral
presentations were made, mostly focussing on both
science and management of nutrient enrichment and
eutrophication. The papers in this Special Issue of
Hydrobiologia are all based on presentations made at
the Symposium.
About the symposium
The symposium focused on the following four topics:
(1) new and existing knowledge regarding coastal
eutrophication, (2) specific eutrophication issues such
as: (a) definition(s) and causes, (b) nutrient cycling
and nutrient limitation, (c) reference conditions and
(d) linkages to other pressures (climate change and
top/down control), (3) summaries of existing knowl-
edge in relation to monitoring and modelling coastal
eutrophication and (4) adaptive environmental man-
agement strategies in relation to coastal
eutrophication.
The symposium was jointly organised by the
Danish Environmental Protection Agency (EPA), the
Swedish EPA, Fyn County and DHI Water &
Environment and received financial support from
the organising institutions. In addition, the Sympo-
sium has been kindly sponsored by: (1) Baltic Sea
2020, (2) Danish Agriculture, (3) the International
Agency for 14C Determination, (4) MARE—the
Swedish Marine Eutrophication Research Programme
and (5) the University of Southern Denmark. Further,
the symposium received support from the European
Commission’s Joint Research Centre, Hotel Nyborg
Strand and Scandinavian Airlines Systems (SAS).
The planning of the symposium was coordinated
by an Organising Committee with the overall
responsibility and a Scientific Committee which
compiled a broad programme focussing on both
science and management. A list of members of the
Electronic supplementary material The online version ofthis article (doi:10.1007/s10750-009-9758-0) containssupplementary material, which is available to authorized users.
Guest editors: J. H. Andersen & D. J. Conley
Eutrophication in Coastal Ecosystems: Selected papers from
the Second International Symposium on Research and
Management of Eutrophication in Coastal Ecosystems, 20–23
June 2006, Nyborg, Denmark
J. H. Andersen (&)
DHI, Agern Alle 5, 2970 Hørsholm, Denmark
e-mail: [email protected]
D. J. Conley
Department of Geology, GeoBiosphere Science Centre,
Lund University, Solvegatan 12, 223 62 Lund, Sweden
123
Hydrobiologia (2009) 629:1–4
DOI 10.1007/s10750-009-9758-0
committees is available as supplementary online
material.
Eutrophication research and management—the
Danish connection
The symposium was a follow up to the highly
successful 1993 Symposium Nutrient Dynamics in
Coastal and Estuarine Environments, organised by
the Danish EPA in collaboration with the European
Commission, Directorate-General for Science,
Research and Development. The Symposium Pro-
ceedings were published in the journal Ophelia with
several seminal papers, for example, Duarte (1995),
Nixon (1995) and Richardson & Heilmann (1995).
There was great regional and international interest
for a follow-up symposium with a focus on both
science and management. This interest in science and
management has been stimulated by legislative
settings, particularly the EU Water Framework
Directive, in which coastal eutrophication problems
are important issues in adaptive management plans
(Anon., 2000).
During recent decades, Denmark and Sweden have
been at the forefront of research on and management
of eutrophication in coastal marine ecosystems
(Jørgensen & Richardson, 1996; Christensen et al.,
1998; Carstensen et al., 2006; Table 1), partly
because the straits between Denmark and Sweden
connecting the Baltic Sea to the North Sea are
vulnerable to nutrient enrichment. Denmark and
neighbouring countries have made substantial efforts
to improve the marine environment through nutrient
reductions both at the national level and through
decades of regional cooperation regarding the Baltic
Sea under the Helsinki Convention (www.helcom.fi)
and the North Sea through the OSPAR Convention
(www.ospar.org).
Both Denmark (Fig. 1) and Sweden have made
large reductions in the discharge of nutrients. Billions
of Euros have been spent, and they have not been
spent in vain. The point source inputs of nutrients to
the marine environment are significantly lower than
they were 20 years ago. However, these reductions
have not been sufficient to reduce the harmful effects
of eutrophication and the targets for improved
ecological status have not been reached.
Three Danish Action Plans for the Aquatic Envi-
ronment over the past two decades (Conley et al.,
2002) have resulted in significant reductions in the
loss of nutrients to the environment (Conley et al.,
2002; Carstensen et al., 2006). Point source inputs of
phosphorus have decreased by more than 80%.
Losses of nitrogen are expected to be reduced by
*50% when changes in agricultural practises that
have already been implemented result in reduced
loads to the marine environment. Figure 1 shows the
temporal trends in total nitrogen loading to the
Kattegat and Danish Straits over a 100-year period,
with a peak in total nitrogen loading in the 1980s.
Since the late 1970s, loads originating from both
point sources and diffuse sources have been declin-
ing. However, more than three decades since the first
measures were implemented and more than a decade
after the First International Danish Symposium on
eutrophication, the problems associated with eutro-
phication are still far from being resolved. There has
been a major development in scientific knowledge
and in the conceptual understanding of nutrient
enrichment and eutrophication in coastal waters.
New questions and challenges have emerged, espe-
cially in relation to modelling and management of
coastal eutrophication. In parallel, new legal and
management settings have emerged or will emerge in
the near future, for example, the EU Water Frame-
work Directive and the process in relation to the
implementation of the European Marine Strategy.
Therefore, it was proposed and agreed in 2004 that a
follow-up symposium focussing on both science and
management of coastal eutrophication should be
organised for June 2006.
About this Special Issue
The 21 papers in this Special Issue are a mixture of
Research Papers, Opinion Papers and Short Notes,
which reflect the broad range of presentations at the
June 2006 symposium. Each manuscript was
reviewed by at least two independent reviewers and
by one of the guest editors. Copy editing was
conducted by Janet F. Pawlak and Carolyn Symon.
The Special Issue has received direct financial
support from the Nordic Council of Ministers via
the working group on the Sea and Atmosphere
2 Hydrobiologia (2009) 629:1–4
123
(Hav- og Luftgruppen) as well as indirect financial
support from the Danish EPA, Swedish EPA and DHI.
This Special Issue as well as others (Kononen &
Bonsdorf, 2001; Rabalais & Nixon, 2002; Bachmann
et al., 2006) demonstrate that considerable knowledge
has been generated since the First Danish Symposium
in 1993. We, as guest editors, are pleased with the
Special Issue as compiled and hope that the readers
will share this opinion.
Despite the vast knowledge and common under-
standing of eutrophication, some important gaps still
remain, especially with regard to regime shifts,
thresholds and multiple stressors. In addition, climate
change needs to be taken into account. A fundamental
problem that needs to be addressed is the lack of
political will to implement adequate nutrient man-
agement strategies. A broader acceptance of the need
to use the best scientific information we have (whilst
still seeking to improve knowledge = ‘‘moving
whilst improving’’) rather than wait for ‘perfection’
is recommended. Finally, it should be kept in mind
that we do not manage eutrophication as such, we
manage humans with the aim of reducing the effects
of eutrophication.
Acknowledgements Thanks are expressed to the Nordic
Council of Ministers and the organisers and sponsors of the
Second International Symposium on Research and
Management of Eutrophication in Coastal Ecosystems. The
Preface improved as a result of comments from Jacob
Carstensen and Scott W. Nixon. Special thanks go to Sif
Johansson, Jørgen Dan Petersen, Jens Brøgger Jensen, Torkil
Jønch Clausen, Henning Karup, Jørgen Magner and Morten
Søndergaard. We are indebted to the reviewers and to Janet F.
Pawlak and Carolyn Symon; this Special Issue would not have
been possible but for their helping hands.
Table 1 Danish nutrient reduction targets sensu the Action Plans for the Aquatic Environment I, II and III. Baseline is 1987;
reductions and targets were agreed by the Danish Parliament in 1987 and subsequently adjusted in 1990, 1999 and 2004
Sector Total nitrogen loads (tonnes) Total phosphorus loads (tonnes)
1987 7 Reduction % = Target 1987 7 Reduction % = Target
Agriculture 311,000 7 152,400 49 = 158,600 4.400 7 4,000 91 = 400
UWWTPs 18,000 7 11,400 63 = 6,600 4.470 7 3,250 73 = 1,220
Industries 5,000 7 3 60 = 2,000 1.250 7 1,050 84 = 200
Total 334,000 7 166,800 50 = 167,200 10.120 7 8,300 82 = 1,820
See Carstensen et al. (2006) for details
UWWTPs: urban wastewater treatment plant effluents
0
20
40
60
80
100
120
140
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
To
tal n
itro
gen
inp
ut (
106 k
g)
Diffuse sourcesPoint sources
Fig. 1 Trends in estimated
total nitrogen inputs (solidline) from Denmark to the
Danish Straits including the
Kattegat since 1900, with 5-
year averages of point and
diffuse sources. Used with
the kind permission of
Jacob Carstensen, NERI;
based on Conley et al.
(2007)
Hydrobiologia (2009) 629:1–4 3
123
Open Access This article is distributed under the terms of the
Creative Commons Attribution Noncommercial License which
permits any noncommercial use, distribution, and reproduction
in any medium, provided the original author(s) and source are
credited.
References
Anon., 2000. Directive 200/60/EC of the European parliament
and of the council of 23 October 2000 establishing a
framework for community action in the field of water
policy. Official Journal L 327/1.
Bachmann, R. W., J. E. Cloern, R. E. Hecky & D. W. Schindler
(eds), 2006. Eutrophication in freshwater and marine
systems. Limnology & Oceanography 51: 351–800
(Special Issue).
Carstensen, J., D. J. Conley, J. H. Andersen & G. Ærtbjerg,
2006. Coastal eutrophication and trend reversal: a Danish
case study. Limnology & Oceanology 51: 398–408.
Christensen, P. B., F. Møhlenberg, L. C. Lund-Hansen, J.
Borum, C. Christiansen, S. E. Larsen, M. E. Hansen, J. H.
Andersen & J. Kirkegaard, 1998. The Danish marine
environment: has action improved its state? Havforskning
fra Miljøstyrelsen, Nr. 62. 115 pp.
Conley, D. J., S. Markager, J. Andersen, T. Ellermann & L. M.
Svendsen, 2002. Coastal eutrophication and the Danish
National Aquatic Monitoring and Assessment Program.
Estuaries 25: 706–719.
Conley, D. J., J. Carstensen, G. Ærtebjerg, P. B. Christensen, T.
Dalsgaard, J. L. S. Hansen & A. Josefson, 2007. Long-
term changes and impacts of hypoxia in Danish coastal
waters. Ecological Applications 17: 165–184.
Duarte, C. M., 1995. Submerged aquatic vegetation in relation
to different nutrient regimes. Ophelia 41: 87–112.
Jørgensen, B. B. & K. Richardson (eds), 1996. Eutrophication
in Coastal Marine Ecosystems. Coastal and Estuarine
Studies, 52. American Geophysical Union, Washington,
DC: 273 pp.
Kononen, K. & E. Bonsdorf (eds), 2001. Man and the Baltic
Sea. Ambio 30: 171–326 (Special Issue).
Nixon, S. W., 1995. Coastal marine eutrophication: a defini-
tion, social causes, and future concerns. Ophelia 41: 199–
219.
Rabalais, N. N. & S. W. Nixon (eds), 2002. Nutrient over-
enrichment in coastal waters: global patterns of cause and
effect. Estuaries 25: 639–900 (Dedicated Issue).
Richardson, K. & J. P. Heilmann, 1995. Primary production in
the Kattegat: past and present. Ophelia 41: 317–328.
4 Hydrobiologia (2009) 629:1–4
123
Marine Pollution Bulletin 60 (2010) 919–924
Contents lists available at ScienceDirect
Marine Pollution Bulletin
journal homepage: www.elsevier .com/locate /marpolbul
Note
A simple method for confidence rating of eutrophication status classifications
Jesper H. Andersen a,*, Ciarán Murray a, Hermanni Kaartokallio b, Philip Axe c, Jarle Molvær d
a DHI Water � Environment � Health, Department of Ecology and Environment, Agern Allé 5, DK-2970 Hørsholm, Denmarkb Finnish Environment Institute, Marine Research Centre, Helsinki, Finlandc Swedish Meteorological and Hydrological Institute (SMHI), Västra Frölunda, Swedend Norwegian Institute for Water Research (NIVA), Oslo, Norway
a r t i c l e i n f o
Keywords:EutrophicationClassificationIndicatorsHEATConfidence ratingWater Framework Directive
0025-326X/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.marpolbul.2010.03.020
* Corresponding author. Tel.: +45 4516 9235.E-mail address: [email protected] (J.H. Andersen).
a b s t r a c t
We report the development of a methodology for assessing confidence in ecological status classifications.The method presented here can be considered as a secondary assessment, supporting the primary assess-ment of eutrophication or ecological status. The confidence assessment is based on scoring the quality ofthe indicators on which the primary assessment is made. This represents a first step towards linking sta-tus classification with information regarding their accuracy and precision. Applied to an existing data setused for assessment of eutrophication status of the Baltic Sea (including the Kattegat and Danish Straits)we demonstrate that confidence in the assessment is Good or High in 149 out of 189 areas assessed (79%).Contrary to our expectations, assessments of the open parts of the Baltic Sea have a higher confidencethan assessments of coastal waters. We also find that in open waters of the Baltic Sea, some biologicalindicators have a higher confidence than indicators representing physical–chemical conditions. In coastalwaters, phytoplankton, submerged aquatic vegetation and indicators of physical–chemical conditionshave a higher confidence than indicators of the quality of benthic invertebrate communities. Our analysesalso show that the perceived weaknesses of eutrophication assessments are due more to Low confidencein reference conditions and acceptable deviations, rather than in the monitoring data.
� 2010 Elsevier Ltd. All rights reserved.
1. Introduction
The purpose of this article is to present for the first time a meth-odology for rating the confidence of classifications of marine eutro-phication status. The evaluation and testing of the presentedmethodology for the confidence assessment has been conductedwith the dataset used in the recent HELCOM integrated thematicassessment of eutrophication in the Baltic Sea (HELCOM, 2009).The results provide valuable information and show a possibleway forward for the assessment of confidence in future environ-mental status assessments.
The three key terms discussed in the present paper are: accuracy,precision, and confidence. If accuracy is the degree of closeness of ameasured or calculated quantity to its actual (true) value, precisioncan be considered to be the degree of reproducibility of that mea-surement. Calculations or measurements can be accurate thoughnot precise, precise but not accurate, neither, or both. A measure-ment system or computational method is considered valid if it isboth accurate and precise. Valid methods inspire confidence. Accu-racy and precision are statistical terms, when we use the word con-fidence on its own, it has a non-statistical meaning.
ll rights reserved.
Numerous assessments of eutrophication status are being pro-duced in Europe, ranging from local assessments to national andregional ones. Nowadays local assessments are tightly linked tothe European Water Framework Directive (WFD). Other types ofassessments can be found in EEA (2001), Ærtebjerg et al. (2003),OSPAR (2008) and HELCOM (2009). The majority of the assess-ments are perceived to be based on sound science and the use ofindicators, but relatively few are based on the use of multi-metricassessment tools. Only few classification tools in accordance withthe WFD exist (e.g. Borja et al., 2009), but most European countriesare developing or testing indicators and/or tools that can be used inthe future.
Currently, the most widely used multi-metric assessment toolsfor assessing marine eutrophication in Europe are those of the OS-PAR Comprehensive Procedure (see OSPAR, 2008; Claussen et al.,2009) and the HELCOM Eutrophication Assessment Tool (HEAT)(see HELCOM, 2009; Andersen et al., submitted for publication).To our knowledge, only one assessment of marine eutrophicationstatus has included an attempt at confidence assessments: the re-cently published HELCOM integrated thematic assessment ofeutrophication in the Baltic Sea region (HELCOM, 2009). The maincause of the apparent lack of this form of secondary assessment isthe limited used of multi-metric indicator-based assessmenttools.
Table 1Scoring matrix. A score of 3 indicates the best possible classification where a ‘‘High”quality rating has been assigned to RefCon, AcDev and to AcStat. A score of 9 indicatesthe worst possible classification, where all scores are rated ‘‘Low”. The indicatorscores from 3 to 9 are converted to an indicator confidence rating ranging from 0 to100%. The worst possible result is assigned a confidence score of 0% and the bestpossible score is equivalent to a confidence score of 100%. The confidence score forother combinations is arrived at by interpolating linearly.
ScoresP
Ind_conf
RefCon AcDev AcStat
1 1 1 3 100%1 1 2 4 83%1 2 2 5 67%2 2 2 6 50%2 2 3 7 33%2 3 3 8 17%3 3 3 9 0%
920 J.H. Andersen et al. / Marine Pollution Bulletin 60 (2010) 919–924
2. Materials and methods
2.1. Data and information sources
This work is based on the existing data used in the HELCOMassessment of Baltic Sea eutrophication 2001–2006 (HELCOM,2009). Most of the monitoring data representing actual status orig-inate from the HELCOM Cooperative Monitoring in the Baltic Mar-ine Environment (HELCOM COMBINE) Programme although somecome from national monitoring and assessment activities. HEL-COM COMBINE is a cooperative monitoring programme sharedby the Baltic countries. Andersen et al. (submitted for publication)describes the data origin in more detail.
2.2. Primary assessment: Eutrophication status
The assessment tool used to illustrate how confidence ratingcan be done is the HELCOM Eutrophication Assessment Tool(HEAT). This is a multi-metric indicator-based tool developed forthe assessment and classification of the eutrophication status ofthe entire Baltic Sea. A total of 189 areas were assessed using indi-cators where information on reference conditions (RefCon) andacceptable deviation from reference conditions (AcDev) could becombined with national monitoring data describing the actual sta-tus (AcStat) for the period 2001–2006 (Andersen et al., submittedfor publication). Using the described RefCon, AcDev, and AcStatconcepts, the basic assessment principle is:
EutroQO ðindicatorÞ ¼ RefCon� AcDev ð1Þ
where EutroQO is an ‘‘eutrophication quality objective” (or target),RefCon is an ‘anchor’ for the assessment while AcDev is the‘yardstick’.
For indicators which have a positive response to nutrient inputs,the classification is determined by the following:
If AcStat < RefConþ AcDev; then the EutroQO ðor targetÞ is met
ð2Þ
Similarly for indicators having a negative response to nutrientinputs:
If AcStat > RefCon� AcDev; then the EutroQO ðor targetÞ is met
ð3Þ
The HEAT tool integrates the elements described above and isbased on: (1) indicators representing well documented eutrophica-tion effects with synoptic information on reference conditions(RefCon), acceptable deviations (AcDev) and actual status (AcStat),(2) quality elements sensu the EU Water Framework Directive (seeAnon., 2000), (3) HELCOM Ecological Objectives (see HELCOM,2009), (4) the relative weighting of indicators within quality ele-ments, and (5) integration of the quality elements used into a finalassessment based on the ‘One out – all out’ principle sensu theWater Framework Directive.
The primary assessment calculated by HEAT is a classification ofeutrophication status in five classes: High, Good, Moderate, Poor,and Bad. The EutroQO (or target) corresponds to the boundary be-tween Good and Moderate status. Assessment results are describedin HELCOM (2009) and Andersen et al. (submitted for publication).
2.3. Secondary assessment: Confidence rating
HEAT also produces an overall confidence rating for each indica-tor, where scores are assigned to each RefCon, AcDev, and AcStatvalue. This scoring is based on expert judgment, where the qualityof RefCon, AcDev, and AcStat is assigned to one of three classes:High (score = 1), Good (score = 2), and Low (score = 3). High and
Good are considered acceptable, while Low indicates a problem re-lated to the quality of the input parameters. We acknowledge thatthe system has a degree of subjectivity since it relies on expertjudgment. Details of the scoring principles are described in Section2.4.
A so-called Final Confidence Rating (FCR) is then calculated inthree steps: after combining the RefCon, AcDev, and AcStat scoresinto an Interim Indicator confidence (see Table 1), then the qualityelement interim confidence (QE-IC) is calculated by taking theweighted arithmetic mean of the confidences of the indicatorswithin the quality element (QE). The Final Confidence Rating(FCR) for a station/water body is then obtained from the arithmeticmean of quality element interim confidences (QE-IC). In calculatingthe FCR, the quality elements are weighted equally, though qualityelements not having any indicators are ignored. For example, at astation where only two quality elements have indicators, the finalconfidence is arrived at by giving a weighting of 50% to each ofthese two QE’s. For each station, separate Interim Confidence rat-ings are also calculated for RefCon, AcDev, and AcStat (respectivelyRefCon-IC, AcDev-IC, AcStat-IC) by taking the arithmetic mean ofthe values for all indicators from all quality elements, withoutany weighting. All HEAT spread sheets including the secondaryassessment of confidence can be found in the Electronic Supple-mentary material.
Experiences from the HEAT classifications have lead to a princi-ple where the final classification of eutrophication status has to bebased on at least two, but preferably at least three QEs, with ideallya minimum of two indicators per QE (data not shown). This hasbeen incorporated in two ways. Firstly, a QE with only one indica-tor has its QE-IC reduced by 25%. Secondly, if the assessment isbased on only a single QE, its FCR is reduced by 50%.
The FCR has three quality classes: High (100–75%), Good (75–50%), and Low (50–0%). This is comparable to the method usedfor analysis of Data Completeness and Reliability (DCR) in the AS-SETS tool developed for assessment of eutrophication status ofestuaries in the United States (NOAA, 2007). In HEAT, High andGood confidence ratings are considered acceptable, while Lowindicates a problem related to the quality of the eutrophicationclassification.
2.4. Scoring principles
Initially, indicators used to assess open water eutrophicationstatus were provisionally scored by a group of national experts(six persons) from Finland (lead country), Denmark, Germany, Po-land and Sweden. For coastal waters, a provisional scoring wasmade by 2–3 national experts from each of the Baltic Sea countries.
J.H. Andersen et al. / Marine Pollution Bulletin 60 (2010) 919–924 921
The type of data used for setting RefCon values, whether it behistorical data, modelling or expert judgment (HELCOM, 2009),has implications for the scoring. In many cases, historical or mod-elled data are used directly, especially if the data are already pub-lished or if the methods are in line with the monitoring methodsused for AcStat. Because RefCons are commonly based on a combi-nation of methods, e.g. (1) historical data and statistical modelling,(2) historical data and dynamical modelling, or (3) historical dataand expert judgment, the level of confidence in the RefCon has tobe taken into account.
The sources of information on AcDevs, and hence the scoring,differ slightly for open and coastal waters. For open waters, wegenerally use +50% and �25%, but other values are used if justified(HELCOM, 2006; OSPAR, 2008; HELCOM, 2009; Andersen et al.,submitted for publication). For coastal waters, we use AcDevs orig-inating from the WFD implementation process as far as possible(e.g. Anon., 2008a, and also Andersen et al., 2004; Krause-Jensenet al., 2005; Henriksen, 2009; Lysiak-Pastuszak et al., 2009a,b,c).If no coast-specific AcDevs are available, we use the AcDevs de-rived for open waters (see HELCOM, 2006; Andersen et al., submit-ted for publication). These different approaches are taken intoaccount where scoring the indicators, since coastal AcDevs in mostcases are better justified and documented compared to those foropen waters. In addition, AcDevs larger than +50% and �25% arein general considered outside the range of minor or slight devia-tions from RefCons and are therefore given a lower confidencescore.
The scoring of AcStat, where the information sources are mostlyin situ monitoring (and in a few cases also modelled data) is basedon the reliability of the observations, their spatial coverage and fre-quency. The confidence assessment could also consider the num-ber of years of adequate data, QA/QC procedures, and whetherdata have been reviewed and published. A non-exhaustive list ofissues to consider when scoring the quality of an indicator is pre-
Table 2Key issues of concern and questions to be addressed when scoring an indicator in HEATinfluence on the score; ‘"’ indicates a positive influence; ‘;’ indicate a negative influence;
Indicatorpart
Concerns and questions
RefCon Are reliable historical data used?Is the method used then comparable with today monitoring methodsIs the spatial and temporal coverage for historical data representativAre RefCon data modelled and justified?Are RefCon data from so-called reference/undisturbed sites being useIs the influence of expert judgment low?Are the RefCon values published in a peer reviewed paper or report s
AcDev Open waters:� Is the indicator a HELCOM target indicator?b
� Is the AcDev value lower than +50% (for a positive indicatoconcentration)?� Is the AcDev value lower than �25% (for a negative indicator resp� Is a functional relation established between nutrient concentratio� Is the influence of expert judgement low?� Is the AcDev value published in a peer reviewed paper or report
Coastal waters:� Does the AvDev value originate from the WFD implementation?� Is the AcDev value lower than +50%?� Is the AcDev value lower than �25%?� Is a functional relation established?� Is the degree of expert judgment low?� Is the AcDev value published in a peer reviewed paper or report
AcStat Does the indicator represent eutrophication well?Is the indicator reported in a HELCOM indicator fact sheet?b
Is the spatial coverage adequate?Is the temporal coverage (frequency) adequate and does it match seaIs the uncertainty known and low?Are AcStat data published in a peer reviewed paper or report series?
a RefCon sites do no longer exist within the Baltic Sea.b More information is available via http://www.helcom.fi.
sented in Table 2. The issues of concern and the questions to beconsidered are closely interconnected and they can sometimesoverlap.
In the second round of the confidence assessment, the scoringfor open and coastal waters was tentatively checked by the con-vener of the assessment work. The results were then, as a thirdround, presented, argued and finally agreed collectively to ensurethat each expert’s assessment methodology was consistent, pro-ducing a harmonized, Baltic Sea-wide assessment (see HELCOM,2008). The final and fourth round in the assessment processincluded verification by the convener and final approval bynational contact points (who were not necessarily the expertswho carried out the assessment).
3. Results and discussion
Confidence ratings were made for 189 areas of the Baltic Seaincluding the Kattegat and Danish Straits. These areas consistedof 172 coastal areas and 17 open water bodies. A summary of allconfidence ratings is presented in Table 3.
3.1. RefCon, AcDev, and AcStat
For RefCons, the average confidence score was 0.75, indicatingthat the experts had High confidence in them. Open waters hadan average RefCon confidence of 0.67, while coastal waters hadan average RefCon confidence of 0.76 (Fig. 1). For AcDev, the pat-tern was the same: The average confidence of AcDev was 0.46,while open waters and coastal waters had average AcDev confi-dence of 0.53 and 0.46, respectively. The AcStat confidence wassomewhat higher with an average AcStat confidence of 0.86. Openwaters had an average AcStat confidence of 0.96, while the AcStatconfidence for coastal waters was only 0.85.
calculations of eutrophication status and confidence. ‘""’ indicates a large positiveand ‘;;’ indicates large negative influence.
Yes / No
" or ;? " or ;
e? " or ;" or ;
d?a ;; or """ or ;;
eries? "" or ;
" or ;r response to nutrient concentrations, such as chlorophyll " or ;
onse, such as Secchi depth)? " or ;ns and indicator response? " or ;
" or ;series? "" or ;
" or ;" or ;;" or ;;" or ;" or ;
series? "" or ;
" or ;;"" or ;" or ;;
sonality? " or ;;"" or ;"" or ;
Table 3Rating of confidence in 14 Baltic Sea basins including the Kattegat and the Danish Straits. High and Good are considered acceptable, while Low indicate a problem related to thequality of the eutrophication classification. Information about the eutrophication status of these basins/areas can be found in HELCOM (2009) and Andersen et al. (submitted forpublication).
Basin High Good Low Total
Bothnian Bay and the Quark 0 (0%) 6 (55.4%) 5 (45.5%) 11 (100%)Bothnian Sea 0 (0%) 9 (40.9%) 13 (59.1%) 22 (100%)The Archipelago and Åland Seas 0 (0%) 5 (83.3%) 1 (16.7%) 6 (100%)Baltic Proper, northern parts 0 (0%) 32 (78.0%) 9 (22.0%) 41 (100%)Gulf of Finland 0 (0%) 18 (90%) 2 (10%) 20 (100%)Baltic Proper, Eastern Gotland Basin 1 (11.1%) 7 (77.8%) 1 (11.1%) 9 (100%)Gulf of Riga 0 (0%) 5 (83.3%) 1 (16.7%) 6 (100%)Western Gotland Basin 0 (0%) 17 (85.0%) 3 (15.0%) 20 (100%)Gulf of Gdansk 1 (20.0%) 3 (60.0%) 1 (20.0%) 5 (100%)Bornholm Basin 2 (14.3%) 9 (64.3%) 3 (21.4%) 14 (100%)Arkona Basin 0 (0%) 4 (100%) 0 (0%) 4 (100%)Kiel Bight and Mecklenburg Bight 0 (0%) 5 (100%) 0 (0%) 5 (100%)Danish Straits including the sound 6 (60.0%) 4 (40.0%) 0 (0%) 10 (100%)Kattegat 5 (31.3%) 10 (62.5%) 1 (6.3%) 16 (100%)Total 15 (7.9%) 134 (70.9%) 40 (21.2%) 189 (100%)
922 J.H. Andersen et al. / Marine Pollution Bulletin 60 (2010) 919–924
AcStat and RefCon generally had an average High confidence.We believe there are two reasons for this. Firstly, the monitoringactivities carried out by the Baltic Sea countries through the HEL-COM COMBINE Programme generally hold a high-quality in termsof the methods used, temporal and spatial resolution as well as QA/QC procedures. Secondly, much effort has been put into establish-ing RefCon values, both for open (HELCOM, 2006) and coastalwaters (Anon., 2008a).
The Low average confidence of AcDev, which are probably moredifficult to establish than RefCon, leave room for improvements,especially for coastal waters. Despite Baltic EU Member States hav-ing spent considerable resources on implementation of the WFD,there seems to be an urgent need for improving the boundary set-ting, since this sets the target for having an acceptable or unaccept-able ecological status.
3.2. Quality elements
The QE based on phytoplankton indicators had an average con-fidence of 0.71, indicating Good confidence (Table 1). The indica-tors used for the phytoplankton quality element were mostly,though not exclusively, based on chlorophyll-a. Assessments ofopen waters and coastal waters had an average confidence of0.74 and 0.71, respectively. Submerged aquatic vegetation (SAV)is used only for assessment of eutrophication status of coastalwaters because of depth limitation. The average confidence forSAV was 0.64, indicating Good confidence. For the QE on benthicinvertebrate communities, the average confidence was 0.33. Foropen waters it was 0.54, whilst it was only 0.30 for coastal waters.For the QE including physical–chemical indicators, the averageconfidence was 0.54. Open waters and coastal waters had an aver-age confidence of 0.62 and 0.53, respectively.
For open waters, the ranking of quality elements in terms ofaverage confidence went from phytoplankton (0.74) > physical–chemical conditions (0.62) > benthic invertebrate communities(0.54), all holding a Good confidence. For coastal water the picturewas different, here the rank was phytoplankton (0.71) > sub-merged aquatic vegetation (0.64) > physical–chemical conditions(0.57) > benthic invertebrates (0.30), the very last being a Low con-fidence QE.
To say that one or more quality element(s) is superior to theothers is tempting. However, it should be kept in mind that allhigh-quality assessments of eutrophication status have to be basedon a range of QEs and indicators. Causative factors, as well as pri-mary and secondary eutrophication effects should be included inany assessment, meaning that indicators dealing with nutrient
concentrations, phytoplankton and benthic communities shouldalways be used. The benthic communities to be included in assess-ments of open (deep) waters will in practice be benthic inverte-brates. For coastal waters, both submerged aquatic vegetationand benthic invertebrates should be included. However, based onthis study, it appears that vegetation (0.64) is considered a morereliable quality element than invertebrate fauna (0.30) at least interms of QE-IC. We assume there are two reasons for this: firstly,the monitoring of SAV is mostly straightforward and based onthe depth limits of the dominating SAV species. Secondly, indicescurrently used to assess the status of benthic invertebrate commu-nities in coastal waters are not completely developed and perhapsnot applicable in brackish or near-limnic coastal waters such as theBaltic Sea.
3.3. Areas
A total of 149 out of the 189 areas assessed were rated as havingHigh or Good confidence. Forty areas were rated as having Lowconfidence (Table 3). In the Arkona Basin, Kiel Bight and Mecklen-burg Bight as well as the Danish Straits including The Sound, noneof the areas assessed hold Low confidence.
Discouragingly, some parts of the Baltic Sea had a high propor-tion of areas with Low confidence ratings (see Table 3), e.g. theBothnian Bay (5 out of 11 or 45.5%), and the Bothnian Sea (13out of 22 or 59.1%). The underlying reasons for this are difficultto deduce. Meaningful explanations could be that the Gulf of Both-nia, surrounded by areas of low population density, is normally re-garded as being unaffected or only slightly affected byeutrophication. Hence, despite about 40% of the runoff to the Balticdraining into the gulf, monitoring activities are relatively modest.This may result in a limited number of available indicators, lowsampling frequencies, indicators with inaccurate setting of Ref-Cons, or AcDevs that are outside the range of what is normallyinterpreted as minor or slight deviation from RefCons. Combina-tions of any or all of these are likely to result in a low FCR.
An important result is that none of Baltic Sea basins had moni-toring and assessment programs which averaged High confidence(see Table 3).
For all areas, the rank of the confidence was as follows: the Dan-ish Straits including The Sound (0.75) > Kattegat (0.69) > ArkonaBasin (0.67) > Kiel Bight and Mecklenburg Bight (0.64) > WesternGotland Basin (0.61) > Gulf of Finland (0.59) > Baltic Proper(0.58) > Bornholm Basin and Gulf of Riga (both 0.57) > Gulf ofGdansk (0.55) > Eastern Gotland Basin (0.54) > Archipelago andÅland Seas (0.52) > Bothnian Bay and the Quark (0.50) > Bothnian
Fig. 1. Confidence ratings of open waters (n = 17) and coastal waters (n = 172) in the Baltic Sea in regard to indicators, quality elements and Final Confidence Ratings (FCR).For the indicators, confidence ratings of reference conditions (RefCon-IC), acceptable deviation (AcDev-IC) and actual status (AcStat-IC) are compared. For the quality element,we compare the confidence rating of phytoplankton (PHY-IC), submerged aquatic vegetation (SAV-IC), benthic invertebrate communities (BIC-IC) and physical–chemicalindicators (PC-IC). The horizontal lines in regard to indicators and quality elements represent the boundary between Good and Low confidence. Please note the y-axes differ inthe FCR figures.
J.H. Andersen et al. / Marine Pollution Bulletin 60 (2010) 919–924 923
Sea (0.46). The monitoring activities in the majority of areas canhardly be reduced without compromising the quality of futureeutrophication assessments. The monitoring of the Bothnian Sea,and perhaps also some of the other areas (e.g. coastal parts ofthe northern parts of the Baltic Proper, Gulf of Gdansk and theBornholm Basin), ought to be improved in order to improve anyfuture assessment of eutrophication status in these areas.
Of the 189 areas assessed, 13 were considered to be unaffectedby eutrophication (HELCOM, 2009; Andersen et al., submitted forpublication). However, only four of these were rated as havingHigh or Good confidence. This does not imply that the remainingnine areas are mis-classified, but it could indicate that the qualityof these status classifications can be questioned. False positiveclassifications (areas mis-classified as being affected by eutrophi-cation) are perhaps a more worrying scenario, at least in budgetaryterms, since investment in load reductions might not be scientifi-cally based. Here, it would be prudent to state that the costs of act-ing on incomplete information or knowledge are generallybelieved to be significantly greater than the costs of obtainingthe information.
3.4. Countries
When looking at individual countries (data not shown), it isclear that some countries (Denmark, Estonia, Germany) had bettercoastal monitoring programs (providing data on AcStat) and indi-cators (providing data on RefCon and AcDev values) than others,leading to Good confidence in most places. Finland, Lithuania, Po-
land and Sweden had acceptable programs, but there might beareas of concern such as in the western coastal waters of the Gulfof Bothnia (see Table 3). Poland might also have reason for concernin some lagoons. The FCR’s made for Latvian and Russian coastalwaters indicate monitoring activities below par, mostly causedby insufficient monitoring activities, both in terms of spatial andtemporal coverage as well as numbers of indicators.
None of the Baltic Sea countries had monitoring programswhich averaged High confidence. For coastal waters, the rank ofthe countries was Denmark (0.70) > Estonia (0.65) > Germany(0.64) > Sweden (0.57) > Finland (0.55) > Lithuania (0.54) > Poland(0.51) > Russia (0.47) > Latvia (0.42). The programs in Denmark,Estonia, Finland, Lithuania, Sweden and Poland cannot be reducedwithout compromising the requirements of the WFD.
An interesting result was that the open waters, often beingmonitored collectively by 2–3 countries per station, had a betteraverage confidence (0.66) than coastal waters (0.57). There areprobably several explanations. Firstly, the monitoring of openwaters has been harmonized and coordinated by HELCOM for al-most three decades, leading to a good understanding of the moni-toring system. Secondly, HELCOM requirements have traditionallyput more focus on the shared open waters rather than the nationalcoastal waters. Thirdly, cruises, even in coastal waters, are expen-sive and while monitoring of offshore regions has benefitted fromthe cost-sharing of the cooperative HELCOM monitoring, some Bal-tic Sea states have had difficulty monitoring their local archipela-gos. Furthermore, at least for some parameters, variability(particularly spatially) is higher in coastal than open waters.
924 J.H. Andersen et al. / Marine Pollution Bulletin 60 (2010) 919–924
3.5. Perspectives
The reported method has several uses, the following five beingthe most obvious: – it can be used to implement or strengthennutrient management strategies in the upstream catchment wherethe primary assessment indicates that the ‘downstream’ area inquestion is ‘affected by eutrophication’ and where the FCR is Highor Good, (2) it can be used to improve the scientific knowledge ofthe basis on which load reductions are being decided or it can pre-vent mistaken investments in load reductions where the primaryassessment indicates that the area in question is an ‘area affectedby eutrophication’ but with a low FCR, (3) it can be used to improvesetting of RefCons and AcDevs, e.g. by encouraging applied re-search in regard to boundary/target setting, (4) it can be used toimprove monitoring activities in areas where the current (2001–2006) confidence is low, and (5) it can be used to warn againstunsupported reductions in monitoring activities, particularlywhere a reduction would change the FCR from Good to Low.
The approach could also be useful in regard to assessmentsaccording to the Water Framework Directive (Anon., 2000) andthe Marine Strategy Framework Directive (Anon., 2008b), sinceEU Member States already are required not only to assess ecologi-cal/environmental status of transitional and coastal marine waters,but also provide information about levels of confidence and preci-sion in assessments (Irvine, 2004).
The guiding principle for the use of the scoring results and con-fidence estimates in managing and developing marine environ-mental assessments should be ‘‘To strengthen and/or maintain thatwhich already has an appropriate quality – Improve what does nothave a good quality”. By doing so, we are likely to, in the longerterm, end up with better monitoring programmes, better indica-tors, better environmental targets, better assessments, and ulti-mately better science-based nutrient management strategies.
4. Conclusions
We have shown that confidence rating of marine eutrophication(and other status assessments) can be performed using a practical,indirect and non-statistical approach. The approach should at thisstage be seen as a first step only, paving the way for more sophis-ticated and formal methodologies. We do not yet intend to pro-mote the suggested methodology, but do stress the need forconfidence rating eutrophication status classifications. However,we consider the methodology reported here to be a useful tool inregard to: (1) indirect assessment of the accuracy and precisionof the primary assessment of eutrophication status, (2) future revi-sion of marine monitoring networks as well as (3) implementationof science-based nutrient management strategies, especially thosepursuant to the EU Water Framework Directive and the EU MarineStrategy Framework Directive.
Acknowledgements
This study is funded by Nordic Council of Ministers (CONFIRM).JHA has been partly funded by HELCOM (via the EUTRO-PRO pro-ject). We would like to thank the following colleagues in the HEL-COM EUTRO-PRO project: Mats Blomkvist, Ulrich Claussen, ViviFleming-Lehtinen, Pirkko Kauppila, Aiste Kubiliute, Elisabeth Ly-siak-Pastuszak, Georg Martin, Günther Nausch, Alf Norkko andAnna Villnäs for providing the data on which this study is based.Special thanks to the DHI-NTU Water and Environment Research
Centre in Singapore and to Uwe Brockmann, David Connor, MariaLaamanen and Flemming Møhlenberg for stimulating discussions.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.marpolbul.2010.03.020.
References
Ærtebjerg, G., Andersen, J.H., Hansen, O.S., (Eds.), 2003. Nutrients andEutrophication in Danish Marine Waters. A Challenge for Science andManagement. National Environmental Research Institute, pp. 126.
Andersen, J.H., et al., submitted for publication. Getting the measure ofeutrophication in the Baltic Sea: towards better assessment principles andmethods. Biogeochemistry (in review).
Andersen, J.H., Conley, D.J., Hedal, S., 2004. Palaeo-ecology, reference conditions andclassification of ecological status: the EU Water Framework Directive inpractice. Marine Pollution Bulletin 49, 282–290.
Anon., 2000. Directive 200/60/EC of the European parliament and of the council of23 October 2000 establishing a framework for community action in the field ofwater policy. Official Journal of the European Communities L 327/1.
Anon., 2008a. Commission decision of 30 October 2008 establishing, pursuant todirective 2000/60/EC of the European parliament and of the council, the valuesof the member state monitoring system classifications as a result of theintercalibration exercise. Official Journal of the European Union L 332/20.
Anon., 2008b. Directive 2008/56/EC of the European parliament and of the councilof 17 June 2008 establishing a framework for community action in the field ofmarine environmental policy (Marine Strategy Framework Directive). OfficialJournal of the European Communities L 164/19.
Borja, A., Bald, J., Franco, J., Larreta, J., Muxika, I., Revilla, M., Rodríguez, J.G., Solaun,O., Uriarte, A., Valencia, V., 2009. Using multiple ecosystem components, inassessing ecological status in Spanish (Basque Country) Atlantic marine waters.Marine Pollution Bulletin 59, 54–64. doi:10.1016/j.marpolbul.2008.11.010.
Claussen, U., Zevenboom, W., Brockmann, U., Topcu, D., Bot, P., 2009. Assessment ofthe eutrophication status of transitional, coastal and marine waters withinOSPAR. Hydrobiologia 629 (1), 49–58. doi:10.1007/s10750-009-9763-3.
EEA, 2001. Eutrophication in Europe’s coastal waters. Topic report 7/2001. Availableat http://www.eea.europa.eu/publications/topic_report_2001_7/Topic_Report_7_2001.pdf.
HELCOM, 2006. Development of tools for assessment of eutrophication in the BalticSea. Baltic Sea Environmental Proceedings No 104, Helsinki Commission, pp. 64.Available via http:www.helcom.fi.
HELCOM, 2008. Minutes of the eight meeting of HELCOM project to elaborate theHELCOM Baltic Sea-wide integrated thematic assessment on eutrophication(HELCOM EUTRO-PRO), pp. 14.
HELCOM, 2009. Eutrophication in the Baltic Sea – an integrated thematicassessment of eutrophication in the Baltic Sea region. Baltic SeaEnvironmental Proceedings No. 115B, Helsinki Commission, pp. 148. Availablevia http://www.helcom.fi.
Henriksen, P., 2009. Reference conditions for phytoplankton at Danish WaterFramework Directive intercalibration sites. Hydrobiologia 629 (1), 255–262.doi:10.1007/s10750-009-9767-z.
Irvine, K., 2004. Classifying ecological status under the European Water FrameworkDirective: the need for monitoring to account for natural variability. AquaticConservation: Marine and Freshwater Ecosystems 14, 107–112. doi:10.1002/aqc.622.
Krause-Jensen, D., Greve, T.M., Nielsen, K., 2005. Eelgrass as a bioindicator under theWater Framework Directive. Water Resources Management 19, 63–75.
Lysiak-Pastuszak, E., Krzyminski, W., Lewandowski, L., 2009a. Development of toolsfor ecological quality assessment in the Polish marine areas according to theWater Framework Directive. Part I – Nutrients. Oceanological andHydrobiological Studies 38 (3), 87–99. doi:10.2478/v10009-009-0037-1.
Lysiak-Pastuszak, E., Krzyminski, W., Lewandowski, L., 2009b. Development of toolsfor ecological quality assessment in the Polish marine areas according to theWater Framework Directive. Part II – Chlorophyll-a. Oceanological andHydrobiological Studies 38 (3), 101–112. doi:10.2478/v10009-009-0038-0.
Lysiak-Pastuszak, E., Krzyminski, W., Lewandowski, L., 2009c. Development of toolsfor ecological quality assessment in the Polish marine areas according to theWater Framework Directive. Part III – Secchi depth. Oceanological andHydrobiological Studies 38 (3), 113–124. doi:10.2478/v10009-009-0039-z.
NOAA, 2007. Effects of nutrient enrichment in the nation’s estuaries: a decadeof change, national estuarine eutrophication assessment update. NOAACoastal Ocean Program Decision Analysis Series No. 26. National Centers forCoastal Ocean Science, Silver Spring, MD, pp. 322. Available via http://www.noaa.gov.
OSPAR, 2008. Second integrated report on the eutrophication status of the OSPARmaritime area. Eutrophication Series, OSPAR Commission, pp. 107. Available viahttp://www.ospar.org.
status of the Baltic Sea. Special focus has been on indi-cators in the following groups: (1) phytoplankton, (2) submerged aquatic vegetation, (3) benthic invertebrates, and (4) supporting features, e.g., nutrient concentrations and water transparency. HELCOM has also focused on the development of tools for the assessment of eutrophica-tion status (HELCOM 2006, 2009a). Combining indicators into a fi nal classifi cation of ‘areas unaffected by eutrophi-cation’ and ‘areas affected by eutrophication’ is carried out using the HELCOM Eutrophication Assessment Tool (HEAT, see Section 1.6, HELCOM (2009a) and Andersen et al. (2010b) for details). HEAT calculates the integrated classifi cation of ‘eutrophication status’. HEAT also calcu-lates a secondary assessment of the confi dence in the eutrophication assessment.
To determine the current status of eutrophication in the Baltic marine ecosystem, the conditions at 17 open-water areas and 172 coastal areas were assessed using data col-lected between 2001 and 2006.
All open waters in the basins of the Baltic Sea, including the open parts of the Bothnian Sea, were found to be ‘affected by eutrophication’. The only open-water areas ‘not affected by eutrophication’ included the open waters of the Bothnian Bay and the Swedish parts of the north-eastern Kattegat, the latter being renewed by oxygen-rich Atlantic waters (Fig. 2.4). The open parts of the Bothnian Sea were labelled ‘affected’ due to increased chlorophyll-a concentrations (see HELCOM 2009a for details).
In most of the coastal waters, nutrient concentrations and chlorophyll-a concentrations generally are elevated com-pared to both target values and reference conditions. In most open basins, mussels, clams, crustaceans and other invertebrates living at the sea fl oor are outside the range of what is considered as being in a ‘good status’.
Only 11 out of 172 coastal areas were found to be ‘unaf-fected by eutrophication’; all of these were located in the Gulf of Bothnia. Outside the Gulf of Bothnia, not a single coastal area in the Baltic achieved this status. Thus, all 161 coastal areas assessed outside the Gulf of Bothnia received the classifi cation ‘affected by eutrophication’. The impaired conditions included elevated levels of nutri-ents and chlorophyll-a, loss of submerged aquatic vegeta-tion, as well as periods of oxygen depletion particularly affecting benthic invertebrates.
The accuracy of the classifi cation results was generally good, although there is some room for improvement. This has been documented indirectly by the rating of confi -dence, in which the data on which the classifi cation was based was scored in terms of accuracy. 145 of 189 areas had an acceptable confi dence level, while the remaining 44 areas had low confi dence. Low confi dence is generally
are themselves harmless, in large quantities they cause eutrophication. The nutrients come from our farmlands, homes and gardens, cars, cities and industries. In the sea, the nutrients fi rst foster the production of plank-tonic algae forming algal blooms, which in the worst case are so large and dense that they are visible even to astronauts in space.
This increased production of organic matter often has secondary and drastic negative consequences: the water becomes murkier and less transparent, the sedimen-tation of organic material to the sea fl oor increases, decomposition of organic matter increases and oxygen is consumed, thus depleting the bottom waters of oxygen. Benthic communities such as meadows of submerged aquatic vegetation are deprived of light, and benthic invertebrate communities and fi sh are affected by oxygen depletion, ultimately suffocating (Fig. 2.3).
Over the years, HELCOM has put considerable efforts into monitoring and assessment of the eutrophication
2.2 EutrophicationEutrophication has its roots in Greek: ‘eu’ meaning ‘well’ and ‘trope’ meaning ‘nourished’, but the translation trivializes the impact of this very serious and expensive ecological syndrome gripping the Baltic. Algal blooms, turbid waters, loss of submerged aquatic vegetation, and dead zones spreading on the sea fl oor – the conse-quences of nutrient inputs and nutrient enrichment in the Baltic are manifold. They have changed the structure and functioning of the marine ecosystem and continue to impair our uses of the ecosystem services.
Eutrophication is triggered by excessive amounts of nutri-ents washed into the sea. Although nutrient chemicals
Figure 2.3 Conceptual model of eutrophication. The arrows indicate the interactions between different ecological compartments. A balanced coastal ecosystem in the Baltic Sea is supposedly characterized by: (1) a short pelagic food chain (phytoplankton > zooplankton > small fi sh > large fi sh), (2) natural species composition of plankton and benthic organisms, and (3) a natural distribution of submerged aquatic vegetation. Nutrient enrichment results in changes in the structure and function of marine ecosystems, as indicated with bold lines. Dashed lines indicate the release of hydrogen sulfi de (H2S) and phosphorus, which both occur under conditions of oxygen depletion. Abbreviations: N = nitrogen; P = phosphorus; Si = silicon; DIN = dissolved inorganic nitrogen; DIP = dissolved inorganic phosphorus.16
Large parts of the Baltic marine ecosystem are trapped in a vicious circle that encourages algal blooms, although the inputs of nitrogen and phosphorus to the sea have been reduced in signifi cant amounts since the late 1980s. In fact, the widespread anoxia which facilitates the release of phosphorus from the sea fl oor sediments fuels the growth and blooms of certain planktonic algae that are capable of utilizing dissolved nitrogen (N2) gas. These algae, termed nitrogen (N2) fi xing blue-green algae or cyanobacteria, are capable of fi xing nitrogen dissolved in the surface layers, thus transforming it into a form that can be used by other organisms. Large quantities of nitrogen compounds available for the growth of other planktonic algae are introduced to the ecosystem by cyanobacteria espe-cially during their bloom period in the late summer. This state is sometimes called a state of repressed recovery (Vahtera et al. 2007).
The limited water exchange with the North Sea and the long residence time of water are the main reasons for the sensitivity of the Baltic Sea to eutrophication. High nutrient loads in combination with a long residence time means that nutrients discharged to the sea will remain in the basin for a long time. In addition, the vertical strati-fi cation of the water masses increases the vulnerability of the Baltic Sea to eutrophication. The most important effect of stratifi cation in terms of eutrophication is that it hinders or prevents ventilation and oxygenation of the bottom waters and sediments by vertical mixing of the water, a situation that often leads to oxygen depletion. Furthermore, hypoxia and anoxia worsen the situation by affecting nutrient transformation processes, such as nitrifi cation and denitrifi cation, as well as the capacity of the sediments to bind phosphorus. In the absence of oxygen, reduced sediments release signifi cant quantities of phosphorus to the overlying water.
a consequence of mediocre monitoring activities or the use of too few or low quality indicators or targets. The interim assessment of confi dence is summarized in Figure 2.4, Panel C. The areas with low confi dence are generally found in the southeastern or northern parts of the Baltic Sea.
Assessing the eutrophication status in an integrated manner for the whole Baltic Sea provides a good basis for evaluating the effectiveness of the implementation of the eutrophication segment of the HELCOM Baltic Sea Action Plan. The assessment clearly documents that nutri-ent inputs need to be further reduced, even though the Baltic Sea countries have successfully reduced nutrient inputs to a certain degree (see Section 3.1.7 and HELCOM 2009a). The eutrophication status of the Baltic Sea will only improve if inputs of both nitrogen and phosphorus are signifi cantly further reduced (Conley et al. 2009b, HELCOM 2009a).
Figure 2.4 Panel A: Integrated classifi cation of eutrophication status in the Baltic Sea (see Fig. 2.2 for an explanation of the interpolation method). Areas in green represent ‘areas unaffected by eutrophica-tion’, while areas in yellow, orange and red represent ‘areas affected by eutrophication’, from Andersen et al. (2010a), based on HELCOM (2009a). Large circles
represent assessment sites in open basins and small circles represent coastal assessment sites. Panel B: Summary of the integrated classifi cations of ‘eutrophi-cation status’ presented as the proportion of assess-ment units per sub-basin, from Andersen et al. (2010a), based on HELCOM (2009a). The colour key is same as in Panel A. Panel C: Interim confi dence ratings of the
eutrophication classifi cations presented as the propor-tion of assessment units per sub-basin. Colours: blue represents high confi dence, green represents accept-able confi dence and red represents a low and hence unacceptable confi dence, from Andersen et al. (2010b), based on HELCOM (2009a).
0 % 20 % 40 % 60 % 80 % 100 %
Bothnian Bay (11)
Bothnian Sea (22)
Archipelago and Åland Seas (6)
Northern Baltic Proper (41)
Gulf of Finland (20)
Western Gotland Basin (9)
Eastern Gotland Basin (6)
Gulf of Riga (20)
Gulf of Gdansk (5)
Bornholm Basin (14)
Arkona Basin (4)
Kiel Bight and Mecklenburg Bight (5)
Belt Sea (10)
Kattegat (16)
good badmoderate poor
Panel B
0 % 20 % 40 % 60 % 80 % 100 %
Bothnian Bay (11)Bothnian Sea (22)
Archipelago and Åland Seas (6)Northern Baltic Proper (41)
Gulf of Finland (20)Western Gotland Basin (9)Eastern Gotland Basin (6)
Gulf of Riga (20)Gulf of Gdansk (5)
Bornholm Basin (14)Arkona Basin (4)
Kiel Bight and Mecklenburg Bight (5)Belt Sea (10)
Kattegat (16)
high acceptable low
Panel C
Panel A
17
Getting the measure of eutrophication in the Baltic Sea:towards improved assessment principles and methods
Jesper H. Andersen • Philip Axe • Hermanni Backer • Jacob Carstensen •
Ulrich Claussen • Vivi Fleming-Lehtinen • Marko Jarvinen • Hermanni Kaartokallio •
Seppo Knuuttila • Samuli Korpinen • Aiste Kubiliute • Maria Laamanen •
Elzbieta Lysiak-Pastuszak • Georg Martin • Ciaran Murray • Flemming Møhlenberg •
Gunther Nausch • Alf Norkko • Anna Villnas
Received: 2 July 2009 / Accepted: 2 July 2010 / Published online: 21 July 2010
� The Author(s) 2010. This article is published with open access at Springerlink.com
Abstract The eutrophication status of the entire
Baltic Sea is classified using a multi-metric indicator-
based assessment tool. A total of 189 areas are
assessed using indicators where information on
reference conditions (RefCon), and acceptable devi-
ation (AcDev) from reference condition could be
combined with national monitoring data from the
period 2001–2006. Most areas (176) are classified as
‘affected by eutrophication’ and only two open water
areas and 11 coastal areas are classified as ‘unaf-
fected by eutrophication’. The classification is made
by application of the recently developed HELCOM
Eutrophication Assessment Tool (HEAT), which is
described in this paper. The use of harmonized
J. H. Andersen (&) � C. Murray � F. Møhlenberg
DHI, Agern Alle 5, 2970 Hørsholm, Denmark
e-mail: [email protected]
P. Axe
SMHI, Nya Varvet 31, 42671 Vastra Frolunda, Sweden
H. Backer � S. Korpinen � M. Laamanen
HELCOM, Katajanokanlaituri 6B, 00160 Helsinki,
Finland
J. Carstensen
National Environmental Research Institute (NERI),
Frederiksborgvej 399, 4000 Roskilde, Denmark
U. Claussen
Federal Environment Agency (UBA), Worlitzer Platz 1,
06844 Dessau-Roßlau, Germany
M. Jarvinen
Finnish Environment Institute (SYKE), Jyvaskyla Office,
P.O. Box 35, 40014 Jyvaskyla, Finland
V. Fleming-Lehtinen � H. Kaartokallio �S. Knuuttila � A. Norkko � A. Villnas
Marine Research Centre, Finnish Environment Institute
(SYKE), P.O. Box 140, 00251 Helsinki, Finland
A. Kubiliute
Center of Marine Research, Taikos Av. 26,
91149 Klaipeda, Lithuania
E. Lysiak-Pastuszak
IMGW, Maritime Branch, Waszyngtona 42,
81-342 Gdynia, Poland
G. Martin
Estonian Marine Institute, University of Tartu, Maealuse
10a, 12618 Tallinn, Estonia
G. Nausch
Leibniz Institute for Baltic Sea Research, Seestr. 15,
18119 Rostock, Germany
A. Norkko
Department of Marine Ecology – Kristineberg, University
of Gothenburg, Kristineberg 566, 45034 Fiskebackskil,
Sweden
123
Biogeochemistry (2011) 106:137–156
DOI 10.1007/s10533-010-9508-4
assessment principles and the HEAT tool allows for
direct comparisons between different parts of the
Baltic Sea despite variations in monitoring activities.
The impaired status of 176 areas is directly related to
nutrient enrichment and elevated loads from
upstream catchments. Baltic Sea States have imple-
mented nutrient management strategies since years
which have reduced nutrient inputs. However, eutro-
phication is still a major problem for large parts of the
Baltic Sea. The 2007 Baltic Sea Action Plan is
projected to further reduce nutrient inputs aiming for
a Baltic Sea unaffected by eutrophication by 2021.
Keywords Eutrophication � Baltic Sea �Assessment � HEAT � Nutrients � Ecological status �Nutrient management strategies
Introduction
Nutrient enrichment, leading to large scale eutrophica-
tion problems in the Baltic Sea, is perhaps the single
greatest threat to the Baltic Sea environment (HELCOM
2009). Nutrient enrichment results in an increase in
productivity and undesirable changes in ecosystem
structure and function (Ryther and Dunstan 1971; Nixon
1995; Cloern 2001). The Baltic Sea ecosystem can
cope with moderate increases in eutrophication pres-
sure, but when the limits of ‘normal’ ecosystem structure
and function are exceeded, eutrophication becomes a
problem (Ærtebjerg et al. 2003; Ronnberg and
Bonsdorff 2004; Feistel et al. 2008; HELCOM 2009).
The 2007 Baltic Sea Action Plan (BSAP),
prepared under the Convention for the Protection of
the Baltic Sea Environment, identifies eutrophication
as one of the four main issues to address in order to
improve the environmental health of the Baltic Sea
(HELCOM 2007a). The BSAP sets a strategic goal
related to eutrophication: ‘a Baltic Sea unaffected by
eutrophication’. This is linked to a set of Ecological
Objectives, which correspond to good ecological/
environmental status sensu the European Water
Framework Directive (WFD) and Marine Strategy
Framework Directive (MSFD) (Anon. 2000, 2008a,
b). The ecological objectives associated with eutro-
phication are: (i) concentrations of nutrients close to
natural levels, (ii) natural levels of algal blooms, (iii)
clear water, (iv) natural distribution and occurrence
of plants and animals, and (v) natural oxygen levels.
In the BSAP, the Baltic Sea states acknowledge
that a harmonized approach to assessing the eutro-
phication status of the Baltic Sea is required.
Therefore, the Baltic Sea states performed a Baltic
Sea-wide thematic assessment of eutrophication sta-
tus including development of a tool for integrated
assessment, the HELCOM Eutrophication Assess-
ment Tool (HEAT). Hence, this article describes the
principles and methods of the HEAT tool.
HEAT builds on the OSPAR Common Procedure
developed for assessment and identification of ‘eutro-
phication problem areas’ in the OSPAR convention
area, in particular the North Sea, the Channel, the
Skagerrak and the Kattegat (see OSPAR 2003, 2008).
It also makes use of some of the key assessment
principles of the WFD, e.g. the calculation of an
Ecological Quality Ratio (EQR) and the ‘one out, all
out’ principle (Anon. 2000; Borja et al. 2009). HEAT
arrives at a primary classification of ‘areas affected by
eutrophication’. In addition, HEAT results in a
secondary assessment of the confidence of the primary
assessment, a feature missing in other eutrophication
assessment tools (Andersen et al. 2010). This study
presents the principles and mechanics of the assess-
ment tool and its results when applied to the Baltic Sea.
Methodology
Study area
The Baltic Sea is an inland sea with a surface area of
415,200 km2 and is one of the largest brackish-water
basins in the world. It is commonly divided into several
sub-basins separated by sills, including a transition
zone to the North Sea consisting of the Kattegat and the
Belt Sea (#11–17 in Fig. 1). These sub-areas differ
considerably in several physical characteristics includ-
ing ice cover, temperature, salinity, and residence time
of the water (Lepparanta and Myrberg 2009). Surface
salinity provides an illustrative example: while it is
normally 20–25 in the Kattegat area, it is only 6–8 in
the central Baltic Sea and drops below 2 in the northern
and eastern extremities of the Bothnian Bay and the
Gulf of Finland. As a result the composition of the biota
changes considerably along these gradients (HEL-
COM 2007b; Feistel et al. 2008).
The human population in the catchment is 85
million, and human activities display a similar,
138 Biogeochemistry (2011) 106:137–156
123
distinctive north–south, east–west pattern. Population
density outside main cities varies from more than 100
persons per km2 in the southern and south-western
parts to less than 1 person per km2 in the northern and
north-eastern parts of the catchment area (CIESN &
CIAT 2005). In terms of land use there is a high
proportion of agricultural land in the south-eastern
and south-western parts, while boreal forest, wetlands
and barren areas dominate in the north (Anon. 2001).
The long residence times (Lepparanta and Myrberg
2009) and the strong saline stratification of the water
column, including natural hypoxia in the deep basins
(Conley et al. 2009a), make large parts of the Baltic Sea
sensitive to nutrient enrichment and eutrophication.
Human activities and settlement, including e.g.
agriculture, urban and industrial waste water, energy
production and transport result in greatly increased
loads of nutrients (nitrogen and phosphorus) from the
(relatively large) 1,700,000 km2 catchment area enter-
ing the Baltic Sea (HELCOM 2004; Schernewski and
Neumann 2005; Savchuk et al. 2008; HELCOM 2009).
Data sources
Three types of data are used in this study: (1)
monitoring data for 2001–2006 (in some cases only
2001–2005 or 2001–2004), (2) information on refer-
ence conditions (RefCon), and (3) ‘target levels’
defined as acceptable deviation (AcDev) from
RefCon.
Fig. 1 The Baltic Sea with
location of ‘assessment
units’ in coastal waters (172
units marked with opencircles) and open basins (17
units shown with numberedcircles). Numbers refer to
Table 1. Reproduced with
permission from HELCOM
Biogeochemistry (2011) 106:137–156 139
123
Most of the monitoring data representing actual
status (AcStat) originate from the HELCOM Coop-
erative Monitoring in the Baltic Marine Environment
Programme (HELCOM COMBINE, see HELCOM
(2008) for details and note that the Kattegat is
included under both HELCOM and OSPAR) carried
out in cooperation between the Baltic countries, and
partly from national monitoring and assessment
activities (e.g. Svendsen et al. 2005; OSPAR 2008).
Data representing long-term trends in inputs of
nutrients (nitrogen and phosphorus) to the Baltic Sea
are derived from the HELCOM Fifth Pollution Load
Compilation (HELCOM 2010). All measurements
and analytical methods used as well as quality
assurance procedures are described in details in the
HELCOM COMBINE Manual, Parts A, B and C
(HELCOM 2008).
In this study, specific focus has been placed
on indicators relevant to HELCOM objectives
(HELCOM 2007b; Backer and Leppanen 2008), in
particular nutrients (objective i), chlorophyll-a (objec-
tive ii), water transparency (objective iii), benthic
invertebrates and submerged aquatic vegetation
(SAV) (objective iv). For the description of the AcStat
all Baltic Sea states have used the 2001–2006 period,
except Denmark, which used the period 2001–2005 for
the Kattegat and Great Belt and 2001–2004 for all
other areas.
RefCon
RefCon, which are ‘‘… a description of the biological
quality elements that exist, or would exist, at high
status, that is, with no, or very minor disturbance
from human activities’’ (Anon. 2000) are used to
quantify the degree of disturbance observed in the
environment. Furthermore, they should represent part
of nature0s continuum and must reflect variability.
Three principles for making the concept of RefCon
operational are (1) reference sites, (2) historical data,
and (3) modelling. Expert judgement can be used as a
supplement when spatially based (option 1 and 2),
modelled (option 3) or combinations of 1, 2 and 3 are
not possible. In this study, the RefCon are mostly
based on historical data and modelling, since refer-
ence sites no longer exist in the Baltic Sea and the use
of expert judgement is occasionally less transparent.
The RefCon’s for nutrients (dissolved inorganic
nitrogen (DIN) and dissolved inorganic phosphorus
(DIP)), chlorophyll-a, water transparency (Secchi
depth) and benthic invertebrates in the open parts of
the Baltic Sea, obtained from various sources
described below, are shown in Fig. 2.
For nutrients, chlorophyll-a and water transpar-
ency, RefCon’s are basin specific and mostly based
on historical data (HELCOM 2006; Fleming-Lehti-
nen 2007; Fleming-Lehtinen et al. 2008; Henriksen
2009). Modelled and site-specific RefCon’s have
been used for parts of the Danish Straits (OSPAR
2008). The reference values used are largely in line
with those presented by other sources, e.g. Sanden
and Hakansson (1996), Aarup (2002) and Schernew-
ski and Neumann (2005).
RefCon’s for benthic invertebrate diversity in open
water basins, measured as gamma diversity, i.e. the
average number of species in a sub-basin per year, were
calculated based upon data from 1965 to 2006
(HELCOM 2009). RefCon’s varied by an order of mag-
nitude between the Arkona Basin and the Bothnian Bay
due to the salinity gradient, which constrains species
distributions (Bonsdorff and Pearsson 1999). For the
coastal water assessments, different national indices
have been used; see HELCOM (2009) for details.
For SAV in coastal waters, namely depth distri-
bution of Fucus vesiculosus and Zostera marina,
which constitute monitoring species in coastal waters
only, RefCon’s are based on historical records, e.g.
Reinke (1889), Waern (1952) and von Wachenfeldt
(1975) as well as Bostrom et al. (2003), Martin
(1999), and Krause-Jensen et al. (2003).
AcDev
For the open basins of the Baltic Sea, AcDev values are
set basin-wise for each indicator. Two different
principles are used for setting the AcDev, according
to whether indicators show a positive response
(increasing in value) to increases in nutrient inputs or
a negative response (decreasing in value). For an
indicator showing positive response (e.g. nutrient
concentrations and chlorophyll-a), AcDev has an
Fig. 2 Reference conditions (RefCon) for open areas of the
Baltic Sea. Numbers refer to Fig. 1. For DIN and DIP greybars are winter mean RefCon’s and black bars are winter
maximum RefCon’s. Please note that no data on DIP are
available for area #4, no data on Secchi depth are available for
areas #12, 13, and 16 and no data on benthic invertebrates are
available for areas #4, 6, and 11–17
c
140 Biogeochemistry (2011) 106:137–156
123
Biogeochemistry (2011) 106:137–156 141
123
upper limit of ?50% deviation from RefCon
(HELCOM 2009). Setting AcDev to 50% implies that
low levels of disturbance (defined as less than ?50%
deviation) resulting from human activity are consid-
ered acceptable while moderate (i.e. greater than
?50%) deviations are not considered acceptable for
the body of water in question. However, in exceptional
cases the ?50% AcDev can be exceeded if scientif-
ically justified. For indicators responding negatively to
increases in nutrient input (e.g. Secchi depth and depth
limit of SAV) the AcDev’s have in principle a limit of
-25% (HELCOM 2009), although AcDev’s used for
benthic invertebrates are slightly greater in magnitude,
ranging from -27 to -40% (HELCOM 2009).
Whereas an indicator with positive response can
theoretically show unlimited deviation, indicators
showing negative response have a maximum deviation
of -100% and a deviation of -25% is, in most cases,
interpreted as the boundary between low and moderate
levels of disturbance. These ?50% and -25% ‘‘prin-
ciples’’ are under discussion, but these initial and
pragmatic values are in accordance with the WFD
(Anon. 2000, 2005) and other eutrophication assess-
ment approaches (Bricker et al. 2003; HELCOM 2006;
NOAA 2007; OSPAR 2008; Bricker et al. 2008;
Claussen et al. 2009). The AcDev’s used for the coastal
waters are largely defined by the WFD implementation
process, in particular the WFD intercalibration activity
in the Baltic Sea (Anon. 2008b).
Assessment principles and methods
The methodology used in this study to assess
eutrophication status of a water body, the HEAT, is
based on indicators, grouped according to a prede-
fined manner. The grouping method used follows the
WFD (Anon. 2000, 2005) quality elements (physical–
chemical features, phytoplankton, SAV, benthic
invertebrates) corresponding to HELCOM eutrophi-
cation objectives i, ii, iii (physical–chemical fea-
tures), iv (phytoplankton) and v (SAV & benthic
invertebrates); subsequently combined into a final
classification of ‘eutrophication status’.
Using the described RefCon, AcDev and AcStat
concepts, the basic assessment principle becomes:
RefCon ± AcDev = EutroQO, where the latter is a
‘‘eutrophication quality objective’’ (or target) corre-
sponding to the boundary between good and moder-
ate ecological status. When the AcStat data exceed
the EutroQO or target, the areas in question is
regarded as ‘affected by eutrophication’’ cf. the
BSAP.
Thus, following the basis assessment principle
described above, a selection of indicators with
RefCon and AcDev values turns qualitative goals
like HELCOM’s five eutrophication objectives into
operational targets, on which objective and transpar-
ent assessments of eutrophication status can be based.
While the RefCon’s can be considered the ‘‘anchors’’
of the assessment, AcDev’s from RefCon’s are the
necessary ‘‘yardsticks’’ while AcStat is actual indi-
cator status. The assessment principles used by
HEAT are summarised in Fig. 3.
The HEAT tool integrates all the elements
described above and is based on: (1) Indicators
representing well documented eutrophication effects
with synoptic information on RefCon, AcDevs, and
AcStat, (2) Quality Elements sensu the WFD, (3)
Fig. 3 Illustration of the key assessment principles used in the
HEAT tool. Please note that HEAT combines the principles of
the HELCOM Baltic Sea Action Plan (right side of the figure
representing open waters) with principles from the EU Water
Framework Directive (left side of the figure representing
coastal waters). Fish by courtesy of Peter Pollard, Scottish EPA
142 Biogeochemistry (2011) 106:137–156
123
HELCOM Ecological Objectives, (4) weighting of
indicators within quality elements, and (5) integration
of the Quality Elements used into a final assessment
based on the ‘One out—all out’ principle sensu the
WFD.
Step 1: Indicators and boundary setting
The EQR is a dimensionless measure of the observed
value (AcStat) of an indicator compared with the
reference value (RefCon). The ratio is equal to 1.00 if
AcStat is better than or equal to RefCon and
approaches 0.00 as deviation from RefCon becomes
large.
Step 1A: Indicators with a positive numerical
relationship to nutrient input For an indicator
showing positive response to nutrient input, the
EQR is defined by:
EQR ¼ RefCon=AcStat ð1Þ0�EQR� 1 ð2Þ
where the observed value of the indicator (AcStat) is
equal to or less than the reference value, then the
EQR is equal to the maximum achievable, 1.00. For a
given reference value, increasing values of AcStat
give lower EQR, with EQR approaching zero as the
status value becomes infinitely large (Fig. 4a).
The value of EQR is used to assign a quality class
to the observed status. The classes in descending
order of quality are RefCon, High, Good, Moderate,
Poor, Bad. The central definition of the quality
classes is given by the value of AcDev. The boundary
between Good and Moderate status is defined as
being where the deviation from RefCon is equal to
the AcDev. That is:
AcStat ¼ 1þ AcDevð Þ � RefCon ð3Þ
Substituting for AcStat in (1) gives:
EQRGood=Moderate ¼ 1= 1þ AcDevð Þ ð4Þ
The EQR boundary between High and Reference
status is always set equal to 0.95. If EQR is above
0.95, it is implicitly assumed that the indicator has a
status equal to RefCon. This deviation is allowed in
order to take into account a degree of uncertainty in
the observations of RefCon and present status as well.
Thus, this permissible deviation from RefCon (5%)
represents a generic estimate of the uncertainty
margin for all indicators. The quality class of
‘‘Reference’’ will rarely be used and quality class
‘‘High’’ therefore, in practice, represents the highest
Fig. 4 Illustration of the boundary (target) setting, when the indicator responds numerically positive to nutrient loads and enrichment
(a) and when the indicator responds numerically negative (b)
Biogeochemistry (2011) 106:137–156 143
123
achievable status. However, the High/Ref boundary is
employed in determining boundaries between the
other classes.
The values for the boundary between Reference/
High status and the boundary between Good/Moder-
ate status constitute fixed points from which the
remaining boundary values are calculated. For
practical reasons the span of the two highest classes
and the next two classes have equal width, i.e.:
EQRRefCon=High � EQRGood=Moderate
¼ EQRGood=Moderate � EQRPoor=Bad ð5Þ
That is, the difference between the values of EQR
defining the Reference/High and Good/Moderate
boundaries is equal to the difference between the
Good/Moderate and Poor/Bad boundary values. This
Eq. 10 can be rearranged to give the value for the
boundary between Poor and Bad status:
EQRPoor=Bad ¼ 2EQRGood=Moderate � EQRRefCon=High
ð6ÞFor example, consider a case where the AcDev
from RefCon is 50%. The boundary between Good
and Moderate status is 1/(1 ? 0.5) = 0.667. And
according to (6), the boundary between Poor and Bad
status lies at 0.383 (Fig. 3a).
This leaves two remaining boundaries to be
defined, the boundary between Good and High status
and the boundary between Poor and Moderate Status.
These boundaries are defined as the midpoints
between the two adjacent boundaries:
EQRHigh=Good ¼ 0:5EQRRefCon=High
þ 0:5EQRGood=Moderate ð7Þ
EQRModerate=Poor ¼ 0:5EQRGood=Moderate
þ 0:5EQRPoor=Bad ð8Þ
For the example of AcDev equal to 50% the values
for the High/Good and Moderate/Poor boundaries
equal 0.808 and 0.525, respectively. Figure 3a shows
how the value of EQR for the boundary between the
classes varies with the AcDev from RefCon.
The method used for calculating class boundaries
does not allow for use of AcDev greater than 110%
for indicators with a positive response to nutrient
input, as the Poor/Bad boundary would otherwise
become negative (Fig. 3a). Consequently, it would
therefore become impossible to obtain a ‘‘Bad’’ status
as an EQR cannot be negative, irrespective of the
extent to which the observed status exceeds RefCon.
Step 1B: Indicators with a numerical negative
relationship to nutrient input For an indicator
showing a negative response to nutrient input, e.g.
depth limit of SAV or Secchi depth, the EQR is
defined as:
EQR ¼ AcStat=RefCon ð9Þ0�EQR� 1 ð10Þ
Here, for a given reference value, the EQR is
directly proportional to the observed value, and is
equal to the maximum value of 1.00 if the AcStat
equals or exceeds the reference value.
As for the case of positive response, the AcDev
from RefCon is used to define class boundaries for
classification according to EQR value. Again, the
Good/Moderate boundary lies where the deviation
from RefCon is equal to the AcDev (3).
Using (3) to substitute for AcStat in (9), and
remembering that AcDev is negative, gives:
EQRGood=Moderate ¼ ð1� AcDevÞ ð11Þ
For an AcDev of 50%, the boundary for Good/
Moderate status is 0.5. Figure 4b shows how the class
boundaries vary with the AcDev. Given the value for
the Good/Moderate boundary and the Ref/High
boundary (0.95), the values for the remaining
boundaries are calculated in the same manner as
described above for indicators with a positive
response to nutrient input. Figure 4b is useful in
illustrating the limit on allowable AcDev for an
indicator with negative response. Choosing an AcDev
greater than 52.5% would mean that according to the
previously described method of calculating class
boundaries, the Bad/Poor boundary becomes negative
(Fig. 4b) and it is therefore impossible to arrive at a
classification of Bad, no matter how far from RefCon
the observed status is.
Step 2: Quality elements and final classification
An EQR value and a set of class boundaries are
calculated for each indicator, but the overall status
classification depends on a combination of indicators.
First, indicator EQR values are combined to give an
EQR value for a specific Quality Element (QE), and
144 Biogeochemistry (2011) 106:137–156
123
similarly the indicator class boundaries are combined
to give the class boundaries for the QE. In the
simplest case, where all indicators within a QE have
equal weights, the EQR for the QE is the average of
the indicators’ EQRs within the QE and each QE
class boundary (e.g. Moderate/Good boundary) is
found as the average of the class boundary values for
all indicators representing that specific QE.
Within a QE, it is also possible to assign weighting
factors to indicators according to expert judgement.
The classification of the QE is then given by
comparison of the weighted averages of the EQRs
with the weighted averages of the individual class
boundaries. Thus, the same weighting is applied both
in calculation of the EQR for the specific QE as well
as QE class boundary values.
The lowest rated of the QEs will because of the
‘One out—all out’ principle determine to final status
classification. This principle is employed for two
reasons: (1) all five HELCOM objectives for the open
basins are required to be met independently, and (2)
this principle is stated in the WFD (Anon. 2000) for
assessing ecological status of coastal waters.
Results
Eutrophication status in the Baltic Sea has been
calculated for 189 assessment units: 172 coastal areas
and 17 open water bodies. In the open water areas,
monitoring data was combined into larger areas by
calculation of mean values to give a common status
for an entire sub-basin, whereas the coastal areas
were assessed in smaller scale (Fig. 1). The EQR
values for nutrients, chlorophyll-a, water transpar-
ency and the gamma diversity of benthic inverte-
brates are presented in Fig. 5.
For the open water bodies, 15 out of 17 are
classified as ‘areas affected by eutrophication’. The
results are summarised in Table 1. Only the Bothnian
Bay and the north-eastern part of the Kattegat are
regarded as ‘unaffected by eutrophication’. The
results of the open water body classifications for
nutrients, chlorophyll-a, water transparency, and
Fig. 5 Ecological Quality Ratios (EQRs) calculated for open
water bodies for a Dissolved Inorganic Nitrogen (DIN),
b Dissolved Inorganic Phosphorus (DIP), c Chlorophyll-a,
d Water transparency (as Secchi depth), and e gamma diversity
for benthic invertebrates. Numbers refer to Fig. 1. Please note
that no data on DIP are available for area #4, no data on Secchi
depth are available for areas #12, 13, and 16 and no data on
benthic invertebrates are available for areas #4, 6, and 11–17
b
Biogeochemistry (2011) 106:137–156 145
123
benthic invertebrates are presented in the following
sections. The detailed HEAT classifications for are
available as electronic supplementary material in
Andersen et al. (2010).
Nutrients
The highest DIN concentrations are found in the
Bothnian Bay, which is predominantly P-limited
(Tamminen and Andersen 2007) and therefore DIN
may accumulate to reach levels above those in other
basins (for actual data, see electronically supplemen-
tary material in Andersen et al. 2010). DIN concen-
trations in the Gulf of Finland are also high due to
large fluvial input of nutrients mainly from the Neva
River. For the other basins, DIN winter means vary
between 3 and 4 lmol l-l. The Gulf of Riga and the
Gulf of Finland have the highest TN annual means
(26 and 24 lmol l-l, respectively), which are due to
large riverine discharges to both basins (Fig. 5a). The
other basins have TN levels between 18 and
21 lmol l-l, with the lowest concentrations in the
Danish Straits. From the Baltic Proper to the Danish
Straits, there is a natural decreasing spatial gradient
owing to the mixing with Skagerrak surface water
that generally has lower TN levels.
High DIP winter means are found in the Gulf of Riga
and the Gulf of Finland (0.78 and 0.84 lmol l-l,
respectively) owing to the large influence from riverine
discharges and the upwelling of bottom waters rich in
phosphorus deriving from the Baltic Proper (Pitkanen
et al. 2001). DIP levels in the Bothnian Sea, the
Baltic Proper and the Danish Straits are similar
(0.35–0.47 lmol l-l), whereas DIP concentrations in
the Bothnian Bay are very low (0.06 lmol l-l). These
spatial differences are unaltered for TP, with high
levels in the Gulf of Riga and the Gulf of Finland
(0.70 and 0.85 lmol l-l, respectively), moderate TP
levels in the Baltic Proper and the Danish Straits
(*0.58 lmol l-l) with slightly lower levels in the
Bothnian Sea (0.42 lmol l-l) and substantially lower
in the Bothnian Bay (0.16 lmol l-l).
Table 1 Classification of eutrophication status for 17 open water areas in the Baltic Sea region
No. Area Ecological quality ratio Eutrophication status
PC PP BIC
1 Bothnian Bay 0.729 (H) 0.668 (H) 0.830 (G) Good
2 Bothnian Sea 0.724 (G) 0.508 (P) 0.834 (H) Poor
3 Gulf of Finland 0.468 (P) 0.220 (B) 0.394 (B) Bad
4 Gulf of Riga 0.543 (M) 0.340 (B) – Bad
5 Northern Baltic Proper 0.523 (P) 0.231 (B) 0.000 (B) Bad
6 Western Gotland Basin 0.660 (M) 0.432 (P) – Poor
7 Eastern Gotland Basin 0.610 (M) 0.486 (P) 0.116 (B) Bad
8 SE Gotland Basin, open parts 0.745 (G) 0.400 (P) 0.222 (B) Bad
9 Bornholm Basin 0.602 (M) 0.553 (M) 0.239 (B) Bad
10 Arkona Basin 0.616 (M) 0.535 (M) 0.764 (G) Moderate
11 Great Belt 0.356 (B) 0.295 (B) – Bad
12 Kattegat, south-western 0.716 (H) 0.460 (P) 0.584 (B) Bad
13 Kattegat, south open parts 0.561 (M) 0.351 (B) – Bad
14 Kattegat, south-eastern 0.821 (G) 0.588 (M) – Moderate
15 Kattegat, central 0.691 (M) 0.440 (P) 0.549 (M) Poor
16 Kattegat, north-eastern 0.787 (G) 0.813 (H) – Good
17 Kattegat, north-western 0.845 (H) 0.603 (M) – Moderate
The eutrophication status is based on the ‘One out—all out’ principle. See Fig. 1 for location of the areas. Detailed HEAT
calculations are available as Electronic Supplementary Material in Andersen et al. (2010). Please note that all values are EQR values
Please note that the EQR values in bold are decisive for the final classification of eutrophication status
PC physical–chemical indicators, PP phytoplankton, and BIC benthic invertebrate communities, H High, G Good, M Moderate,
P Poor, B Bad
146 Biogeochemistry (2011) 106:137–156
123
The EQR values for DIN vary between 0.22 and
0.81 (see Fig. 5a). For DIP, EQR values vary
between 0.33 and 1.00, the latter being an indication
of almost pristine conditions in the Bothnian Bay and
the Bothnian Sea (Fig. 5b). As expected, nutrient
status is acceptable in the Bothnian Bay (area 1). The
only other areas where nutrient status is acceptable
are the northern parts of the Kattegat (areas 16 and
17), areas 2 (Bothnian Sea), 8 (south-eastern Baltic
Proper), and 14 (south-eastern Kattegat).
Phytoplankton and water transparency
Mean summer (June–September) chlorophyll-a con-
centrations are highest for the open water bodies in
the Gulf of Finland, the Northern Baltic Proper and
the Gulf of Riga (5.4, 4.8 and 5.3 lg l-l, respec-
tively). In other open water bodies, average chloro-
phyll-a concentrations range from 1.9 to 2.7 lg l-l.
The variability in summer (June–September) chloro-
phyll-a observations in 2001–2006 is high, with
individual values ranging from 0.1 to [50 lg l-l.
In most of the open Baltic Sea areas, chlorophyll-a
concentrations indicate eutrophication. In other
words, EQR values derived for chlorophyll-a show
a clear deviation from RefCon (Fig. 5c). In the open
sea, the chlorophyll-a derived status is the highest in
the Bothnian Bay and the Kattegat (0.67 and 0.63,
respectively) and lowest in the Gulf of Finland, the
Northern Baltic Proper, and the Gulf of Riga (0.22,
0.23 and 0.34, respectively).
Reduced water transparency is partly an effect of
increased nutrient loads, mediated through increased
phytoplankton growth. In comparison to RefCon
(Fig. 5d), water transparency status has decreased in
all Baltic Sea sub-areas at both at coastal and open
sea sites reflecting visible eutrophication effects in
the entire Baltic Sea.
Water transparency status in open sea areas
expressed as EQR values vary markedly in different
sub-basins of the Baltic Sea. Status expressed as EQR
values varies from 0.75 to 1.0 for the southern and
central sub-basins, indicating a 0–25% decrease in
water transparency from near-pristine RefCon. How-
ever, sub-basins north of the Northern Baltic Proper
have a significantly lower status with EQR values
ranging from 0.50 to 0.61, representing a reduction of
39–50% in water transparency compared to RefCon.
The mean EQR value for all open sub-basins assessed
is 0.72. In the south-eastern Gotland Basin and
Arkona Basin water transparency status is highest of
all open sub-basins, with EQR values of 1.0 and 0.94
respectively. In the Kattegat water transparency
status exceeds the mean status (mean EQR for
Kattegat sites 0.75). In the Bornholm Basin, the
Western and Eastern Gotland Basin, the EQR values
are nearly equal to the Kattegat (0.75–0.81). In Gulf
of Riga, the two indicators used for Secchi depth have
variable RefCon (4.0 m for the Finnish indicator and
6.0 m for the Latvian indicator) and result in different
EQR values of 0.75 and 0.57, respectively.
The Northern Baltic Proper and Gulf of Finland
represent a distinctly lower status compared to
RefCon, with EQR values of 0.61 in the open
Northern Baltic Proper and 0.50 in the Gulf of
Finland. In the open sea areas of the Gulf of Bothnia
water transparency EQR is 0.61 in the Bothnian Sea
and 0.56 in the Bothnian Bay.
Benthic invertebrates
No benthic invertebrates survive in areas with
prolonged or permanent oxygen depletion such as in
the deep parts of the Baltic Proper. In areas with
periodic oxygen depletion every late summer and
autumn, the number of benthic species is reduced
significantly and mature communities cannot develop.
In marine areas with temporary oxygen depletion,
intermittent recovery will occur whenever conditions
improve. Oxygen depletion, if rare enough, may be
viewed as a temporal and spatial mosaic of distur-
bance that results in the loss of habitats, reductions in
biodiversity, and a loss of functionally important
species. Macrobenthic communities are severely
degraded throughout the open sea areas of the Baltic
Proper and the Gulf of Finland, whereas conditions in
the Arkona Basin, the Bothnian Sea and Bothnian Bay
are classified as being good (Fig. 5e).
For the open waters, the EQR values vary between
0.00 and 0.83. The highest EQR values are as
indicated above found in the Arkona Basin (0.77), the
Bothnian Sea (0.83) and the Bothnian Bay (0.83). For
the Baltic Proper and the Gulf of Finland, EQR
values range from 0.00 to 0.39 indicating impaired
environmental conditions.
Biogeochemistry (2011) 106:137–156 147
123
Coastal waters
Of the 172 coastal waters assessed, 161 are classified
as ‘affected by eutrophication’ (Table 2). Coastal
waters are in general more vulnerable to nutrient
inputs than open waters—important causes being the
lower retention times as well as closer benthic-
pelagic interactions (Borum 1996; Wasmund et al.
2001). Seasonal variations in supply, removal, and
transformation processes give rise to distinct seasonal
patterns for nutrient concentrations in Baltic Sea
coastal areas. Distinct spatial gradients are also
found, with elevated nutrient concentrations in estu-
aries and coastal waters compared to open waters.
This gradient is most pronounced in the Danish
Straits and Baltic Proper. Nutrient concentrations in
coastal areas of the Gulf of Finland are similar to
those in the open sea because of upwelling of
offshore bottom water. Detailed information on
nutrient status of the coastal waters can be found in
HELCOM (2009) and Lysiak-Pastuszak et al. (2009).
In a majority of coastal Baltic areas, chlorophyll-a
concentrations and water transparency measurements
indicate the prevalence of eutrophication (data not
shown). In other words, EQR values derived from
chlorophyll-a and water transparency measurements
show a clear deviation from RefCon. Detailed
information about the status of planktonic communi-
ties and water transparency in various coastal waters
of the Baltic Sea can be found in Feistel et al. (2008)
and HELCOM (2009).
Extensive seagrass meadows and perennial mac-
roalgal communities harbour the highest biodiversity
in coastal, shallow-water ecosystems. Eutrophication
has complex effects on SAV causing shifting of the
distribution depth limit towards the surface, prevent-
ing the settlement of new specimens on the seafloor
due to increased sedimentation, and favouring oppor-
tunistic species with a short life cycle and rapid
development over the perennial species, thus causing
a shift in community composition. Generally, the
level of eutrophication has caused serious changes in
the Baltic Sea SAV communities, although in many
cases the gaps in historical data do not allow us to
identify the exact timing of larger shifts in commu-
nities (Torn et al. 2006). Present-day monitoring data
Table 2 Summary of eutrophication status classifications of 172 coastal water bodies in the Baltic Sea region
Basins and sub-basins Eutrophication status classification Total
High Good Moderate Poor Bad
Bothnian Bay 0 1 3 2 2 8
The Quark 0 1 1 0 0 2
Bothnian Sea 0 9 6 2 4 21
The Archipelago and Aland Seas 0 0 2 1 3 6
Gulf of Finland 0 0 4 6 9 19
Gulf of Riga 0 0 0 3 2 5
Baltic Proper, northern parts 0 0 3 7 30 40
Eastern Gotland Basin 0 0 0 0 7 7
Western Gotland Basin 0 0 0 5 14 19
Gulf of Gdansk 0 0 0 1 4 5
Bornholm Basin 0 0 1 7 5 13
Arkona Basin 0 0 1 1 1 3
Kiel Bight and Mecklenburg Bight 0 0 0 2 3 5
Danish Straits including the Sound 0 0 1 4 5 10
Kattegat 0 0 3 1 5 9
Total 0 11 25 42 94 172
High and Good represent ‘areas unaffected by eutrophication’, while Moderate, Poor, and Bad represent ‘areas affected by
eutrophication’
148 Biogeochemistry (2011) 106:137–156
123
show that the degradation of communities is ongoing
in several areas (HELCOM 2009). At the same time,
positive signs of a slowing down or reversal of some
eutrophication effects on SAV parameters could be
observed in areas of the Northern Baltic Proper and
the Gulf of Finland, where the previous distribution
of macrophyte species has recovered in some areas
(Nilsson et al. 2004; HELCOM 2009).
In the western part of the Baltic Sea (the Kattegat
and the Danish Straits), the EQR values for the depth
distribution of Zostera marina vary between 0.89 and
0.59. With a -25% AcDev, only the Danish coastal
areas of the Kattegat have average EQR values above
0.75. For the Danish Straits, all average EQR values
are below 0.75, and hence classified as ‘affected by
eutrophication’. In the central, eastern and northern
parts of the Baltic Sea, in areas dominated by Fucus
vesiculosus, average EQR values vary between 0.84
and 0.55. EQR values above 0.75 are found in the
Gulf of Riga and Eastern Baltic Proper. In the
Bothnian Sea, Gulf of Finland, and the western parts
of the Baltic Proper, the targets for SAV are generally
not met.
Macrozoobenthic communities in coastal waters
are highly variable both between and within different
sub-basins. In general, more sheltered and enclosed
coastal water bodies are in a worse state than more
exposed open coasts. Detailed information on status
of benthic invertebrates in Baltic Sea coastal water
can be found in HELCOM (2009).
Integrated assessment
Combining indicators and applying the ‘One out—all
out’ principle in order to produce a final classification
of eutrophication status represents a step forward
from assessments based on individual indicators
towards integrated assessments applying multi-metric
indicator-based assessment tools such as HEAT. The
results can be presented in several ways, e.g.: (1)
HEAT calculations (see electronic supplementary
material in Andersen et al. (2010) for details), (2)
summarised as in Tables 1 and 2 as well as (3) in the
form of maps of eutrophication status in the Baltic
Sea.
Figure 6 presents a merger of HEAT classifica-
tions for 17 open water areas (Table 1) and 172
coastal water bodies (Table 2) into an interpolated
map of eutrophication status of the Baltic Sea. All
open parts of the Baltic Sea except the Bothnian Bay
and the north-eastern parts of the Kattegat are
classified as ‘affected by eutrophication’. It should
be noted that also some coastal waters situated along
the Bothnian Sea are classified as ‘unaffected by
eutrophication’.
Discussion
This assessment of eutrophication status in the Baltic
Sea compares target values (EutroQOs), derived from
combining information on RefCon (representing a
‘then’ situation) and an AcDev with recent
(2001–2006) monitoring data (representing a ‘now’
situation). According to the results of this study only
open parts of the Bothnian Bay and north-eastern
Kattegat as well as some coastal waters in Bothnian
Bay are unaffected by eutrophication.
The results of this study are generally in line with
previous indicator-based assessments (HELCOM
2002, 2006; Ærtebjerg et al. 2003; Ronnberg and
Bonsdorff 2004) and can be directly compared with
the results of national coastal assessments and the EU
processes like WFD implementation in the Baltic
(e.g. Anon. 2008b). An added value of the method
employed here over e.g. WFD is that it uses
supporting parameters, e.g. nutrients and Secchi
depth, which are significantly correlated to the
biological quality elements, on the same level of
importance as the biological quality elements
(Nielsen et al. 2002a, b; Krause-Jensen et al. 2003).
The RefCon values derived for all 17 open water
‘assessment units’ are based on the analysis of
historical data. The RefCon values used for the open
parts of the Baltic Sea represent the best available
knowledge about the eutrophication status of the
Baltic Sea 50–100 years ago before the onset of the
current large scale eutrophication process (Scher-
newski and Neumann 2005; Savchuk et al. 2008) and
the monitoring data used in this study represent the
best available datasets for the area. Hence, these
RefCon values are in principle ready for immediate
use in regard to any updates of the BSAP, e.g. as done
by Wulff et al. 2007.
The principles of this assessment for setting ‘target
values’ (e.g. the AcDev) are in line with the WFD: it
is the boundary between Good Ecological Status and
Moderate Ecological Status according to the WFD
Biogeochemistry (2011) 106:137–156 149
123
(Anon. 2000). For Good Ecological Status, which
together with High Ecological Status, is considered
acceptable status, the values of the biological quality
elements show low levels of disturbance from
RefCon as a result of human activity. For Moderate
Ecological Status, which together with Poor and Bad
Ecological Status, is considered an unacceptable
status, the values of the biological quality elements,
compared to RefCon, deviate moderately (or more)
from those normally associated with the water body
type under undisturbed conditions.
The nutrient concentrations overall reflect the
balance between inputs from land, atmosphere and
loss processes, and are generally in line with other
studies and assessments carried out in the Baltic Sea,
e.g. Lundberg et al. 2009. Nutrient concentrations can
be influenced also by upward mixing from deeper
water layers (Vahtera et al. 2007; Feistel et al. 2008;
Reissmann et al. 2009). Upwelling is an important
source of phosphorus in the Gulf of Finland, the Gulf
of Riga and also in the Baltic Proper (Nausch et al.
2009). The relatively high EQR values found in the
south-eastern Baltic Proper (0.75), the western Got-
land Basin (0.81), and the south-eastern Kattegat
(0.78) are assumed to be related to imprecise setting
of RefCon. There is a need for the development of
more harmonised information on RefCon values for
nutrient concentrations.
Phytoplankton is perhaps the most important
element in any assessment of eutrophication in the
Baltic Sea, since phytoplankton primary production
and biomass are essentially coupled to nutrient
concentrations. Chlorophyll-a concentrations are
widely used as a proxy for phytoplankton biomass,
Fig. 6 Integrated and
interpolated five-class
classification of
eutrophication status in the
Baltic Sea region. The
interpolation was made by
inverse distance weighting
method and the gradients
among the point values
were permitted to change
over intermediate distances.
While the status of offshore
areas was pooled to a single
value from multiple point
values, the coastal
assessment units were
treated as separate and were
given a 25 km effect radius.
Reproduced with
permission from HELCOM
150 Biogeochemistry (2011) 106:137–156
123
but other indicators should be developed, e.g. in
regard to algal species indicative of nuisance or toxic
algal blooms. The findings presented here are gener-
ally in line with other studies and assessments, e.g.
Jaanus et al. (2007), Fleming-Lehtinen et al. (2008),
Hakansson and Lindgren (2008), Wasmund and
Siegel (2008). During recent decades, chlorophyll-a
concentrations have been increasing in most of the
Baltic Sea sub-regions, although in the 2000s chlo-
rophyll-a levels in many open sea areas show signs of
a decreasing trend. RefCon values for chlorophyll-a
in open waters seem appropriate for the time being.
For coastal waters there seem to be a need for joint
principles and methods of setting not only RefCon
values, but also AcDev’s. This has not yet been
achieved by the WFD intercalibration activity.
The assessment of water transparency is closely
linked to the assessment of phytoplankton and SAV,
and in this study water transparency is regarded as a
proxy of eutrophication. An added value in regard to
water transparency is the length of the time series,
which extends close to 100 years back in time
(Sanden and Hakansson 1996). The findings pre-
sented here are generally in line with other studies
and assessments, e.g. Kautsky et al. (1986), and
Eriksson et al. (1998, 2002). In the Gulf of Riga, low
status is consistent with lower RefCon compared to
other areas. Low status in the Gulf of Bothnia may be
attributed mostly to changes in land use affecting
water colour (humic substances), whereas in the Gulf
of Finland the increase of phytoplankton biomass is a
more likely proximate reason for the low status.
The benthic invertebrate assessment for open
waters shows that the benthic communities are
structured by a combination of physical factors (e.g.
salinity and sediment type) and eutrophication, which
result in a higher susceptibility to hypoxia/anoxia.
The findings presented here are generally in line with
other studies and assessments, e.g. Karlson et al.
(2002), and Perus and Bonsdorff (2004). A special
challenge is the difficulty in defining historical
RefCon for macrozoobenthos—this emphasizes the
importance of conducting long-term monitoring over
large spatial scales to be able to assess changes.
Assessment of SAV in coastal waters is, at least
compared to the assessment of open waters, some-
what more challenging because the status of SAV
communities depends on a variety of local environ-
mental conditions which also affect also the
eutrophication processes on very limited, local scale,
e.g. changes in nutrient loading to specific river basin
or fjord or bay while open sea indicators reflect
situation on larger sea area. So it is no surprise that
especially in case of extensive archipelago areas
some SAV indicators can show development in
opposite direction than indicators of nearby open
sea areas. In our case some recovery in the depth
distribution of SAV has occurred during last decades
in the Northern Baltic Proper (extensive archipelago
areas) as well as in some areas of the Gulf of Finland,
while indicators used for open sea areas still show
declining status.
There is in our opinion no such thing as a perfect
assessment tool. More targeted monitoring and
improved understanding of the eutrophication pro-
cesses will lead to better knowledge, better indicators
and subsequently better assessment tool. The strength
of HEAT compared to the OSPAR equivalent on
which it is built, is that it is modernized in the sense it
makes use of (1) the EQR and the ‘one out, all out’
principle. Hence, HEAT is directly linked to the
principles for assessment of ecological status of
coastal water sensu the WFD. An added value
of HEAT is that it enables a secondary assessment
of confidence (see Andersen et al. 2010). Compared
to OSPAR COMP, the HEAT tool has no or few
weaknesses. When using HEAT for assessment of
‘ecological status’ sensu the WFD, it can be argued
that ‘eutrophication status’ and ‘ecological status’ are
different issues. This point is for somewhat mean-
ingless, at least for the Baltic Sea, where the major
threat to the coastal ecosystems is nutrient enrich-
ment and eutrophication. It can also be argued the
combination of indicators per QE mixes indicators
with different boundary setting, but here it should be
eminent that the classes used by the WFD are related
to QE (cf. Annex 5), not to individual indicators or
indices.
By providing a regional overview of eutrophica-
tion status in the Baltic Sea the results of this study
provide interesting perspectives and links to the
implementation of a range of EU Directives, e.g. the
WFD, the MSFD (Anon. 2008a), the EC Urban
Wastewater Treatment Directive (Anon. 1991a) and
the EC Nitrates Directive (Anon. 1991b). The
relations in regard to boundary setting and classifi-
cation are discussed and outlined in Anon. (2009) and
HELCOM (2009). If the convergence of the aims of
Biogeochemistry (2011) 106:137–156 151
123
these directives is taken seriously, marine waters
classified as ‘affected by eutrophication’ could by no
means be accepted as having either ‘Good Ecological
Status’ or habitats with a ‘Favourable Conservation
Status’. Similarly it can be argued that waters
classified as ‘affected by eutrophication’ should be
designated as ‘sensitive’ to nutrient inputs from
industries and cities. Along the same lines waters
affected by eutrophication should be regarded as
‘polluted’ when situated downstream of catchment
dominated by agriculture, implying that the catch-
ment should be designated as ‘vulnerable’ in regard
to losses of nitrogen from agricultural practices.
Future assessments will however be worthless if
we fail to safeguard the current spatial and temporal
resolution of HELCOM COMBINE and monitoring
for the joint HELCOM core set of eutrophication
indicators. Any weakening of these activities will
jeopardize future re-assessments of eutrophication
status of the Baltic Sea. Issues to be improved before
a re-assessment include: (1) harmonization and
evaluation of the quality of reference condition
values (RefCon), (2) improvements of the target
values (e.g. AcDev) (more research on functional
relations, natural variations etc.), (3) improvements in
spatial and temporal coverage of HELCOM COM-
BINE monitoring in some areas (e.g. Gulf of Riga,
eastern Baltic Proper, South-eastern Baltic proper),
(4) adequate monitoring of SAV, (5) development of
oxygen indicators, and (6) development of statistical
principles for weighting indicators.
The current impaired status of most parts of the
Baltic Sea is a consequence of a combined increase in
population density and altered agricultural practices.
This has resulted in increased discharges, emissions
(including atmospheric nitrogen emissions) and
losses of nutrients to the environment and ultimately
nutrient enrichment in the aquatic environment. Only
few data series of nutrient loading exist, e.g.
Stalnacke (1996) and Conley et al. (2007), and
hence, the long-term nutrient enrichment will have to
be documented by the temporal trends for TN and TP
concentrations as well as TN:TP ratio in surface
(0–10 m) and bottom waters ([100 m) starting from
the 1970s until 2006 (HELCOM 2009).
Nutrient concentrations increased until the 1980s,
and in all areas except for the Gulf of Finland,
phosphorus concentrations have declined during the
past two decades (HELCOM 2009). Nitrogen
concentrations have declined in the Gulf of Riga,
the Baltic Proper and the Danish Straits. These
declines, particularly in the coastal zone, are partly
caused by lower nutrient loads from land. Further-
more, changing volumes of hypoxia in the Baltic
Proper significantly alter nutrient concentrations in
bottom waters and, through subsequently mixing,
also in surface waters. This does not affect the Baltic
Proper alone but also connecting basins through
advective exchanges. In particular, the Gulf of
Finland has been severely affected by internal
loading of phosphorus from the sediments caused
by poor oxygen conditions (Vahtera et al. 2007).
The elevated nutrient concentrations compared to
RefCon are primarily a consequence of a long-term
(100? years) increase in direct and riverine loads to
the Baltic Sea. However, management strategies
focusing mainly on direct discharges have during
the last 20 years resulted in a decrease in loads to the
Baltic Sea (Fig. 7). However, it has to be taken into
Fig. 7 Trends in inputs of total nitrogen (TN) and total
phosphorus (TP) to the Baltic Sea. Please note that the TP input
has been scaled by factor 10. The solid line indicate run off in
m3 s-1. ‘‘2003’’ = 2001–2003 and ‘‘2006’’ = 2001–2006.
2021 (grey bars) show the ultimate nutrient input targets to
be reached as agreed by the HELCOM Baltic Sea Action Plan
152 Biogeochemistry (2011) 106:137–156
123
account that decreased flow is also partly responsible
for decreasing loads (HELCOM 2009).
Improving the eutrophication status, especially of
those areas classified as affected by eutrophication,
relies on a better linking of ecosystem effects,
nutrient concentrations, loads and human activities
in upstream catchments. The key issue is to reverse
the trend of eutrophication, sometimes referred to as
oligotrophication (Nixon 2009), and to reduce inputs
of nutrients to the Baltic Sea region. Some improve-
ment has been made in some regions (Carstensen
et al. 2006 and Fig. 7) but additional reductions are
clearly needed. Recent modelling efforts (Wulff et al.
2007; Savchuk et al. 2008) have come a long way in
providing advice on the magnitude of nutrient input
reductions required to reach identified target levels of
key parameters, such as those utilised in this study
(Fig. 2). A first round of such calculations was
actually adopted in 2007 by Baltic Sea states in the
BSAP, partly based on an ecosystem approach to
management of human activities (HELCOM 2007b;
Wulff et al. 2007). Recently, a process of revision of
these reduction figures was begun, taking into
account more assessment parameters and atmospheric
deposition to better reflect relevant ecosystem ele-
ments and all relevant pathways of nutrient input.
When developing and implementing ecosystem-
based nutrient management strategies, it has been
debated whether a nutrient management strategy such
as the BSAP should focus either on N, P or both
(Tamminen and Andersen 2007). Given the varia-
tions in nutrient limitation between region and
seasons—and the fact that the flow out of the Baltic
Sea passes areas which are nitrogen limited—it is
clear that alleviation of eutrophication requires a
balanced and strategic approach to control both
nitrogen and phosphorus appropriately (Conley
et al. 2009b).
What we consider in our assessment of eutrophi-
cation or ecological status being a straightforward
eutrophication signal is in reality a response not only
to nutrient enrichment, but also to many other
pressures (Jackson et al. 2001). Often the functional
relations are complicated, including issues like
thresholds, regime shifts and climate change (Duarte
et al. 2009; Duarte 2009). The implications for
management are currently being understood and
interpreted. A rational solution would be to acknowl-
edge that other pressures (e.g. climate change) might
enhance eutrophication signals and that further efforts
in regard to reduction of nutrient inputs may been
needed to comply with most eutrophication related
objectives.
Conclusions
This study has introduced a multi-metric indicator-
based eutrophication assessment tool enabling a
harmonized assessment of eutrophication status in
the whole Baltic Sea. Most parts of the Baltic Sea are,
not surprisingly, judging from available scientific
literature, affected by nutrient enrichment.
The recently developed HEAT as described in this
paper provides a qualified answer to this key question
‘‘Do we have a problem or not?’’ and thus a basis for
the implementation or revision of a Baltic Sea-wide
nutrient management strategy, e.g. the BSAP.
HEAT represents a major step forward in terms of
assessing eutrophication in the Baltic Sea. Firstly,
because HEAT is based on well-established eutro-
phication indicators, it is in line with the principles of
the WFD, and, perhaps more importantly, it uses the
EQR approach to enable direct comparisons of all
areas assessed despite variation in monitoring activ-
ities. Secondly, HEAT classifications can be regarded
as a baseline for the reduction figures defined in the
eutrophication segment of the BSAP against which
the HELCOM vision of a Baltic Sea unaffected by
eutrophication can be judged.
HEAT has shown itself to be a good tool and
should be used for a HELCOM re-assessment of the
eutrophication status of the Baltic Sea within e.g.
6–10 years in order to follow the implementation of
the BSAP and validate the effectiveness of the
reduction measures established so far.
Future assessments should be based on the best
scientifically based indicators and assessment tools
available rather than waiting for so-called ‘perfec-
tion’. However, development of eutrophication
assessment tools and nutrient management strategies
in the Baltic and elsewhere should ideally be
adaptive: there should always be the intention to
adapt these tools when new scientific knowledge
becomes available. Similarly, nutrient management
strategies should be based on the best available
science-based functional relations between causes
and effects, using models and Decisions Support
Biogeochemistry (2011) 106:137–156 153
123
Systems as appropriate. Eutrophication in the Baltic
Sea is a significant challenge and the absence of
faultless tools should not prevent the Baltic Sea
countries from trying to meet this challenge.
Acknowledgements The views expressed are those of the
authors and do not necessarily represent official positions of
their institutions. The authors would like to thank national and
local authorities, and colleagues for making monitoring data
available for the HELCOM integrated thematic assessment of
eutrophication in the Baltic Sea region. Special thanks are owed
to: Juris Aigars, Mats Blomqvist, Saara Back, Alf B. Josefson,
Henning Karup, Pirkko Kauppila, Pirjo Kuuppo, Juha-Markku
Leppanen, Barbel Muller-Karulis, Samuli Neuvonen, Janet
Pawlak, Heikki Pitkanen, Johnny Reker and Roger Sedin. This
work has received financial support from HELCOM (HELCOM
EUTRO and HELCOM EUTRO-PRO), the Danish Ministry for
the Environment (CO-EUTRO) and DHI.
Open Access This article is distributed under the terms of the
Creative Commons Attribution Noncommercial License which
permits any noncommercial use, distribution, and reproduction
in any medium, provided the original author(s) and source are
credited.
References
Aarup T (2002) Transparency of the North Sea and Baltic
Sea—a Secchi depth data mining study. Oceanologia
44:323–337
Ærtebjerg G, Andersen JH, Hansen OS (eds) (2003) Nutrients
and eutrophication in Danish marine waters. A challenge
for science and management. National Environmental
Research Institute, Roskilde, p 126
Andersen JH, Murray C, Kaartokallio H, Axe P, Molvær J
(2010) A simple method for confidence rating of eutro-
phication status classifications. Mar Pollut Bull. doi:
10.1016/j.marpolbul.2010.03.020
Anon. (1991a) Council Directive of 21 May 1991 concerning
urban waste water treatment (91/271/EEC). Official
Journal of the European Communities L 135
Anon. (1991b) Council Directive 91/676/EEC of 12 December
1991 concerning the protection of waters against pollution
caused by nitrates from agricultural sources. Official
Journal of the European Communities L 375
Anon. (2000) Directive 200/60/EC of the European Parliament
and of the Council of 23 October2000 establishing a
framework for Community action in the field of water
policy. Official Journal of the European Communities L
327/1
Anon. (2001) Land cover for Baltic Sea drainage basin region.
http://www.grida.no/baltic/htmls/related1.htm. Accessed
19 May 2009
Anon. (2005) Common Implementation Strategy for the Water
Framework Directive (2000/60/EC) Guidance Document
No. 14. Guidance on the Intercalibration Process
2004–2006, 26 pp
Anon. (2008a) Directive 2008/56/EC of the European Parlia-
ment and of the Council of 17 June 2008 establishing a
framework for community action in the field of marine
environmental policy (Marine Strategy Framework
Directive). Official Journal of the European Communities
L 164/19
Anon. (2008b) Commission decision of 30 October 2008
establishing, pursuant to Directive 2000/60/EC of the
European Parliament and of the Council, the values of the
Member State monitoring system classifications as a result
of the intercalibration exercise. Official Journal of the
European Union L 332/20
Anon. (2009) Guidance document on eutrophication assess-
ment in the context of European water policies. Version
14, 17 May 2009, 140 pp
Backer H, Leppanen JM (2008) The HELCOM system of a
vision, strategic goals and ecological objectives: imple-
menting an ecosystem approach to the management of
human activities in the Baltic Sea. Aquat Cons 18(3):
321–334
Bonsdorff E, Pearson TH (1999) Variation in the sublittoral
macrozoobenthos of the Baltic Sea along environmental
gradients: a functional group approach. Aust J Ecol
24:312–326
Borja A, Bald J, Franco J, Larreta J, Muxika I, Revilla M,
Rodrıguez JG, Solaun O, Uriarte A, Valencia V (2009)
Using multiple ecosystem components, in assessing eco-
logical status in Spanish (Basque Country) Atlantic mar-
ine waters. Mar Pollut Bull 59:54–64
Borum J (1996) Shallow waters and land/sea boundaries. In:
Jørgensen BB, Richardson K (eds) Eutrophication in
coastal marine ecosystems. Coastal and Estuarine Studies
52. American Geophysical Union, Washington, DC, p 272
Bostrom C, Baden SP, Krause-Jensen D (2003) The seagrasses
of Scandinavia and the Baltic Sea. In: Green EP, Short FT
(eds) World atlas of seagrasses. University of California
Press, Berkeley, pp 27–37
Bricker SB, Ferreira JG, Simas T (2003) An integrated meth-
odology for assessment of estuarine status. Ecol Modell
169:39–60
Bricker S, Longstaff B, Dennison W, Jones A, Boicourt K,
Wicks C, Woerner J (2008) Effects of nutrient enrichment
in the nation’s estuaries: a decade of change. Harmful
Algae 8:21–32
Carstensen J, Conley DJ, Andersen JH, Ærtebjerg G (2006)
Coastal eutrophication and trend reversal: a Danish case
study. Limnol Oceanogr 51(1,2):398–408
CIESN & CIAT (2005) Center for International Earth Science
Information Network (CIESIN), Columbia University;
and Centro Internacional de Agricultura Tropical (CIAT).
Gridded Population of the World Version 3 (GPWv3):
Population Density Grids. Palisades, NY: Socioeconomic
Data and Applications Center (SEDAC), Columbia Uni-
versity. http://sedac.ciesin.columbia.edu/gpw. Accessed
19 May 2009
Claussen U, Zevenboom W, Brockmann U, Topcu D, Bot P
(2009) Assessment of the eutrophication status of transi-
tional, coastal and marine waters within OSPAR. Hydro-
biologia. doi:10.1007/s10750-009-9763-3
Cloern J (2001) Our evolving conceptual model of the coastal
eutrophication problem. Mar Ecol Prog Ser 210:223–253
154 Biogeochemistry (2011) 106:137–156
123
Conley DJ, Carstensen J, Ærtebjerg G, Christensen PB,
Dalsgaard T, Hansen JLS, Josefson AB (2007) Long-term
changes and impacts of hypoxia in Danish coastal waters.
Ecol Appl 17:165–184
Conley DJ, Bjorck S, Bonsdorff E, Carstensen J, Destouni G
et al (2009a) Hypoxia-related processes in the Baltic Sea.
Environ Sci Technol 43(10):3412–3420
Conley DJ, Paerl HW, Howarth RW, Boesch DF, Seitzinger
SP, Havens KE, Lancelot C, Likens GE (2009b) Con-
trolling eutrophication: nitrogen and phosphorus. Science
323:1014–1015
Duarte CM (2009) Coastal eutrophication research: a
new awareness. Hydrobiologia. doi:10.1007/s10750-009-
9795-8
Duarte CM, Conley DJ, Carstensen J, Sanchez-Camacho M
(2009) Return to neverland: shifting baselines affect
eutrophication restoration targets. Estuar Coasts
32:29–36. doi:10.1007/s12237-008-9111-2
Eriksson BK, Johansson G, Snoeijs P (1998) Long-term
changes in the sublittoral zonation of brown algae in the
southern Bothnian Sea. Eur J Phycol 33:241–249
Eriksson BK, Johansson G, Snoeijs P (2002) Long-term
changes in the macroalgal vegetation of the inner Gullmar
Fjord Swedish Skagerrak coast. J Phycol 38:284–296
Feistel R, Nausch G, Wasmund N (eds) (2008) State and
evolution of the Baltic Sea, 1952–2005. John Wiley &
Sons Inc, Hoboken, New Jersey, p 703
Fleming-Lehtinen V (ed) (2007) HELCOM EUTRO: devel-
opment of tools for a thematic eutrophication assessment
for two Baltic Sea sub-regions, The Gulf of Finland and
The Bothnian Bay. MERI No. 61, Report Series of the
Finnish Institute of Marine Research, 35 pp
Fleming-Lehtinen V, Laamanen M, Kuosa H, Haahti H,
Olsonen R (2008) Long-term development of inorganic
nutrients and chlorophyll a in the open Northern Baltic
Sea. Ambio 37:86–92
Hakansson L, Lindgren D (2008) On regime shifts and budgets
for nutrients in the open Baltic proper: evaluations based
on extensive data between 1974 and 2005. J Coast Res
24:246–260
HELCOM (2002) Fourth Periodic Assessment of the State of
the Marine Environment of the Baltic Sea, 1994–1998.
Baltic Sea Environment Proceedings No. 82B. Helsinki
Commission, 215 pp
HELCOM (2004) The Fourth Baltic Sea Pollution Load
Compilation (PLC-4). Baltic Sea Environment Proceed-
ings No. 93. Helsinki Commission, 188 pp
HELCOM (2006) Development of tools for assessment of
eutrophication in the Baltic Sea. Baltic Sea Environmental
Proceedings No. 104. Helsinki Commission, 64 pp
HELCOM (2007a) HELCOM Baltic Sea Action Plan. Helsinki
Commission, 103 pp
HELCOM (2007b) Pearls of the Baltic Sea. In: Hegerhall
Aniasson B, Hagerhall B (eds) Networking for life: spe-
cial nature in a special sea. Helsinki Commission, 198 pp
HELCOM (2008) Manual for Marine Monitoring in the
COMBINE Programme of HELCOM. http://www.
helcom.fi/groups/monas/CombineManual/en_GB/main/.
Accessed 4 June 2009
HELCOM (2009) Eutrophication in the Baltic Sea—an inte-
grated thematic assessment of eutrophication in the Baltic
Sea region. Baltic Sea Environmental Proceedings No.
115B. Helsinki Commission, 148 pp
HELCOM (2010) Fifth Periodic Load Compilation. Helsinki
Commission (in press)
Henriksen P (2009) Reference conditions for phytoplankton at
Danish Water Framework Directive intercalibration sites.
Hydrobiologia. doi:10.1007/s10750-009-9767-z
Jaanus A, Andersson A, Hajdu S, Huseby S, Jurgensone I,
Olenina I, Wasmund N, Toming K (2007) Shifts in the
Baltic Sea summer phytoplankton communities in
1992–2006. HELCOM Indicator Fact Sheets 2007.
http://www.helcom.fi/environment2/ifs/en_GB/cover/
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA,
Botsford LW, Bourque BJ, Bradbury RH, Cooke R,
Erlandson J, Estes JA, Hughes TP, Kidwell S, Lange
CB, Lenihan HS, Pandolfi JM, Peterson CH, Steneck
RS, Tegner MJ, Warner RR (2001) Historical overf-
ishing and the recent collapse of coastal ecosystems.
Science 293:629–638
Karlson K, Rosenberg R, Bonsdorff E (2002) Temporal and
spatial large-scale effects of eutrophication and oxygen
deficiency on benthic fauna in Scandinavian and Baltic
waters—a review. Oceanogr Mar Biol 40:427–489
Kautsky N, Kautsky H, Kautsky U, Waern M (1986) Decreased
depth penetration of Fucus vesiculosus (L.) since the
1940’s indicates eutrophication of the Baltic Sea. Mar
Ecol Prog Ser 28:1–8
Krause-Jensen D, Pedersen MF, Jensen C (2003) Regulation of
eelgrass (Zostera marina) cover along depth gradients in
Danish coastal waters. Estuaries 26:866–877
Lepparanta M, Myrberg K (2009) Physical oceanography of
the Baltic Sea. Springer, Berlin, Heidelberg, p 410
Lundberg C, Jakobsson B-M, Bonsdorff E (2009) The
spreading of eutrophication in the eastern coast of the
Gulf of Bothnia, northern Baltic Sea—an analysis in time
and space. Estuar Coast Shelf Sci 82:152–160
Lysiak-Pastuszak E, Krzyminski W, Lewandowski L (2009)
Development of tools for ecological quality assessment in
the Polish marine areas according to the Water Frame-
work Directive. Part I—nutrients. Oceanol Hydrobiol
Stud 38(3):87–99. doi:10.2478/v10009-009-0037-1
Martin G (1999) Distribution of phytobenthos biomass in the
Gulf of Riga (1984–1991). Hydrobiologia 393:181–190
Nausch M, Nausch G, Lass HU, Mohrholz V, Nagel K, Siegel
HM, Wasmund N (2009) Phosphorus input by upwelling
in the eastern Gotland Basin (Baltic Sea) in summer and
its effects on filamentous cyanobacteria. Estuar Coast
Shelf Sci 83(4):434–442. doi:10.1016/j.ecss.2009.04.031
Nielsen SL, Sand-Jensen K, Borum J, Geertz-Hansen O
(2002a) Depth colonisation of eelgrass (Zostera marina)
and macroalgae as determined by water transparency in
Danish coastal waters. Estuaries 25:1025–1032
Nielsen SL, Sand-Jensen K, Borum J, Geertz-Hansen O
(2002b) Phytoplankton, nutrients and transparency in
Danish coastal waters. Estuaries 25:930–937
Nilsson J, Engkvist R, Persson L-E (2004) Long-term decline
and recent recovery of Fucus populations along rocky
shores of southeast Sweden, Baltic Sea. Aquat Ecol 38:
587–598
Nixon SW (1995) Coastal marine eutrophication: a definition,
social causes, and future concerns. Ophelia 41:199–219
Biogeochemistry (2011) 106:137–156 155
123
Nixon SW (2009) Eutrophication and the macroscope. Hyd-
robiologia. doi:10.1007/s10750-009-9759-z
NOAA (2007) Effects of nutrient enrichment in the nation’s
estuaries: a decade of change, national estuarine eutro-
phication assessment update. NOAA Coastal Ocean Pro-
gram Decision Analysis Series No. 26. National Centers
for Coastal Ocean Science, Silver Spring, MD, p 322
OSPAR (2003) OSPAR Integrated Report 2003 on the Eutro-
phication Status of the OSPAR Maritime Area Based
Upon the First Application of the Comprehensive Proce-
dure. Eutrophication Series. OSPAR Commission, 59 pp
OSPAR (2008) Second OSPAR Integrated Report on the
Eutrophication Status of the OSPAR Maritime Area.
Eutrophication Series. OSPAR Commission, 107 pp
Perus J, Bonsdorff E (2004) Long-term changes in macro-
zoobenthos in the Aland archipelago, northern Baltic Sea.
J Sea Res 52:45–56
Pitkanen H, Lehtoranta J, Raike A (2001) Internal nutrient
fluxes counteract decreases in external load: the case of
the estuarial eastern Gulf of Finland, Baltic Sea. Ambio
30:195–201
Reinke J (1889) Algenflora des westlischen Ostsee Deutchen
Antheils. Eine systematisch-pflanzengeographische Stu-
die. Schmidt & Klaunig, Kiel, p 101
Reissmann J, Burchard H, Feistel R, Hagen E, Lass HU,
Mohrholz V, Nausch G, Umlauf L, Wieczorek U (2009)
State of-the-art report on Baltic vertical mixing and con-
sequences for eutrophication. Progr Oceanogr 82:47–80.
doi:10.1016/l.pocean.2007.10.004
Ronnberg C, Bonsdorff E (2004) Baltic Sea eutrophication:
area-specific ecological consequences. Hydrobiologia
514:227–241
Ryther JH, Dunstan WM (1971) Nitrogen, phosphorus, and
eutrophication in the coastal marine environment. Science
171:1008–1013
Sanden P, Hakansson B (1996) Long term trends in Secchi
depth in the Baltic Sea. Limnol Oceanogr 41(2):346–351
Savchuk OP, Wulff F, Hille S, Humborg C, Pollehne F (2008)
The Baltic Sea a century ago—a reconstruction from
model simulations, verified by observations. J Mar Sys
74(1–2):485–494
Schernewski G, Neumann T (2005) The trophic state of the
Baltic Sea a century ago: a model simulation study. J Mar
Sys 53(1–4):109–124
Stalnacke P (1996) Nutrient loads to the Baltic Sea. Linkoping
Studies in Art and Science No. 146, PhD Thesis. Linko-
ping University
Svendsen LM, van der Bijl L, Boutrup S, Norup B (eds) (2005)
NOVANA: Nationwide Monitoring and Assessment Pro-
gramme for the Aquatic and Terrestrial Environments.
Programme Description—Part 2. National Environmental
Research Institute, Denmark. NERI Technical Report No.
537, 137 pp
Tamminen T, Andersen T (2007) Seasonal phytoplankton
nutrient limitation patterns as revealed by bioassays over
Baltic Sea gradients of salinity and eutrophication. Mar
Ecol Prog Ser 340:121–138
Torn K, Krause-Jensen D, Martin G (2006) Present and past
depth distribution of bladderwrack (Fucus vesiculosus) in
the Baltic Sea. Aquat Bot 84:53–62
Vahtera E, Conley DJ, Gustafsson BG, Kuosa H, Pitkanen H,
Savchuk OP, Tamminen T, Viitasalo M, Voss M,
Wasmund N, Wulff F (2007) Internal ecosystem feed-
backs enhance nitrogen-fixing cyanobacteria blooms and
complicate management in the Baltic Sea. Ambio
36:186–194
von Wachenfeldt T (1975) Marine benthic algae and the
environment in the Oresund. I–III. PhD Thesis, Lund
University
Waern M (1952) Rocky-shore algae in the Oregrund archi-
pelago. Acta Phytogeogr Suec 30:1–298
Wasmund N, Siegel H (2008) Chapter 15. Phytoplankton. In: R
Feistel, G Nausch, N Wasmund (eds) State and evolution
of the Baltic Sea, 1952–2005. A detailed 50-year survey
of meteorology and climate, physics, chemistry, biology,
and marine environment. Wiley, pp 441–481
Wasmund N, Andrushaitis A, Lysiak-Pastuszak E, Muller-
Karulis B, Nausch G, Neumann T, Ojaveer H, Olenina I,
Postel L, Witek Z (2001) Trophic status of the south-
eastern Baltic Sea: a comparison of coastal and open
areas. Estuar Coast Shelf Sci 53:849–864
Wulff F, Savchuk OP, Sokolov AV, Humborg C, Morth M
(2007) Management options and effects on a marine
ecosystem: assessing the future of the Baltic Sea. Ambio
36:243–249
156 Biogeochemistry (2011) 106:137–156
123
H
Sa
b
c
a
ARRA
KABCEM
1
eearn(2mdoshvuata
1d
Ecological Indicators 15 (2012) 105–114
Contents lists available at SciVerse ScienceDirect
Ecological Indicators
jo ur nal homep age: www.elsev ier .com/ locate /eco l ind
uman pressures and their potential impact on the Baltic Sea ecosystem
amuli Korpinena,∗, Laura Meskia, Jesper H. Andersenb,c, Maria Laamanena
Helsinki Commission, Katajanokanlaituri 6B, FIN-000160 Helsinki, FinlandDepartment of Bioscience, Aarhus University, Ny Munkegade 120, 8000 Aarhus C, DenmarkNational Centre for Environment and Energy (NERI), Frederiksborgvej 399, 4000 Roskilde, Denmark
r t i c l e i n f o
rticle history:eceived 17 March 2011eceived in revised form 15 August 2011ccepted 19 September 2011
eywords:nthropogenic pressurealtic Seaumulative impactsnvironmental impact assessment
a b s t r a c t
The EU Marine Strategy Framework Directive requires Member States to estimate the level of humanimpacts on their marine waters. We report the first attempt to quantify the magnitude and distributionof cumulative impacts of anthropogenic pressures for an entire regional sea, the Baltic Sea. We useda method which takes account of the sensitivity of different ecosystem components and gives scoresfor potential impacts in 5 km × 5 km areas. Our quantification of impacts was based on data layers ofanthropogenic pressures and ecosystem components. The classification of the anthropogenic pressuresfollows the MSFD and the outcome of the index was targeted to facilitate the implementation of thedirective. The study presents the cumulative impacts over the entire sea area and shows that the highestestimated impacts were in the southern and south-western sea areas and in the Gulf of Finland. The
arine Strategy Framework Directive lowest index values were found in the Gulf of Bothnia. The results coincide with the population densitiesof the adjacent catchment areas. Fishing, inputs of nutrients and organic matter and inputs of hazardoussubstances comprised 25%, 30% and 30%, respectively, of the total cumulative impact. The approach usedis transparent and the results are useful in regard to ecosystem-based management, e.g. for area-basedmanagement and assessments. Examples of uses are given together with analysis of the strengths andweaknesses of the approach.
. Introduction
Human activities place heavy pressures on the global marinecosystems (Millenium Ecosystem Assessment, 2005; Halpernt al., 2008; Crain et al., 2009). Coastal marine environmentsnd marginal seas in particular have experienced ecosystemegime shifts, altered food web structures, heavily contami-ated sediments and adverse effects from hazardous substancesJackson et al., 2001; Kappel, 2005; Casini et al., 2008; Coll et al.,008). A human activity may cause multiple pressures on thearine ecosystem. For example, bottom trawling alone may pro-
uce simultaneously above-water and underwater noise, harvestf target and non-target species, physical disturbance of theea bed, increased siltation, and resuspension of nutrients andazardous substances. Moreover, anthropogenic pressures havearying impacts on different components of the ecosystem whichnderlines the importance of including ecosystem aspects to anssessment of impacts. Despite the long history of human activi-
ies at sea, quantitative spatial analyses of anthropogenic pressuresnd their cumulative impacts on the marine ecosystems have been∗ Corresponding author.E-mail address: [email protected] (S. Korpinen).
470-160X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.ecolind.2011.09.023
© 2011 Elsevier Ltd. All rights reserved.
conducted only recently (Lindeboom, 2005; Halpern et al., 2008,2009; Selkoe et al., 2009; Ban et al., 2010).
Assessment of pressures and impacts is one of the key features ofthe EU Marine Strategy Framework Directive (MSFD, Anon, 2008).The directive requires Member States to make assessments notonly on pressures and impacts but also on the state of the marineenvironment and then take measures towards reaching a goodenvironmental status (GES) by 2020. The MSFD stipulates that GESmeans the environmental status of marine waters where “theseprovide ecologically diverse and dynamic oceans and seas whichare clean, healthy and productive within their intrinsic conditions,and the use of the marine environment is at a level that is sus-tainable, thus safeguarding the potential for uses and activities bycurrent and future generations.” The level of sustainable use is notprescribed in the directive and therefore spatial presentations ofpressures and their potential impacts can be used as a starting pointto iterate such a sustainable level. Despite the tight implementa-tion schedule of the MSFD, no assessment of cumulative impactshas yet been made in any of the European seas.
This study estimates the distribution and magnitude of differ-ent human activities both at sea and on land, associated pressures
and their potential impacts on the marine ecosystem. We used anassessment tool which converts pressures to potential impacts onselected components of the ecosystem and sums up all the impactsin predefined assessment units. The tool is based on the method and
106 S. Korpinen et al. / Ecological Indicators 15 (2012) 105–114
F ities ab
geaad
ig. 1. Map of the Baltic Sea showing the sub-basins, drainage basin, major rivers, clue.
lobal assessment by Halpern et al. (2007, 2008). To our knowl-
dge, this is one of the first spatial visualizations of cumulativenthropogenic pressures and impacts in a coastal sea area glob-lly. The assessment relies on the best available compilation ofata on human activities and ecosystem in the area which has beennd countries. Differences in the water depth are indicated with different shades of
possible to compile due to the long standing regional cooperation of
the Baltic Sea coastal countries and the EU under the umbrella of theHelsinki Commission (HELCOM). This assessment should be seenas the first step towards more comprehensive impact assessmentsand better validated quantification of impacts.
al Indicators 15 (2012) 105–114 107
2
2
ewtwatpeStetb2
2
fapeinTosaoaolhc
af
I
wbsipteamnfiai
2
mp
RESOLUTION: Size of the assessment unit
IDENTIFICATION OF PRESSURES: - Relevant parameter and
unit f or the press ure - Search f or
distributionsand intensity da ta
IDENTIFICATION OF ECOSYSTEM DA TA: - Relevant species, bio-
topes, biotope com-plexes distr ibution etc . - Search for da ta
WEIGHT SCORES : - Involve ex perts - Create guidelines - Crea te a quest ionnaire and
hold a workshop - Take medians of the results
MANAGEMENT PUR POSES : Assessmen ts ( MSFD), Protected areas, (threats) Marine spatial planning, E IAs and permitting, Scenarios
PRESSURE LAYERS: - Log-transformation - Normalization [ 0-1]
ECOSYSTEM DA TA: - 0 / 1 for assessment
units
IND EX CA LCULA TION : - Multiply the t hree fac tors and sum
them up within an assessment uni t
MODIFIED INDICES: Sector or t hemewise index (e.g. only nutrients, fisheries…) Ecosystem component -wise index (e.g. only seagrass or seals ) Index without ecosystem factors
IND
EX
CA
LC
UL
AT
ION
A
PPL
ICA
TIO
NS
S. Korpinen et al. / Ecologic
. Methods
.1. The study area
The Baltic Sea is a semi-enclosed brackish sea area with a gradi-nt in surface water salinity from 31 to 2 PSU and relatively shallowater (average depth 52 m). The oceanic connection is maintained
hrough narrow and shallow Danish straits, from which salineater flows over a series of sills, which separate the Baltic Sea into
series of sub-basins (Fig. 1). Natural features like water residenceime of around 30 years, shallowness and large catchment arearedispose the Baltic Sea to the accumulation of pollution by nutri-nts and hazardous substances. Large cities such as St. Petersburg,tockholm and have resided on the coast for centuries, and overime, anthropogenic pressures have heavily affected the Baltic Seacosystem, which is seen in decreased health of marine top preda-ors, over-exploited fish populations, increased areas of anoxic seaottoms, contaminated fish and extensive algal blooms (HELCOM,009a,b, 2010a).
.2. Measuring cumulative impact
We define an anthropogenic pressure as a human-derived stressactor causing either temporary or permanent disturbance or dam-ge to or loss of one or several components of an ecosystem. Thus,ressure may cause immediate impacts or it may also be lownough not to cause immediate adverse impacts on biota. Accord-ng to our definition, potential anthropogenic impact is the possibleegative change a pressure may cause on an ecosystem component.he impact is only considered potential, because our estimates relyn the current, still imperfect, expert knowledge on the relation-hips between pressures and impacts on the ecosystem and thectual impact can be reduced or increased by natural variability andther stochastic factors. By cumulative impact we mean the sum ofll potential impacts in an area, not taking into account synergisticr antagonistic effects. By an ecosystem component we mean bio-ogical parts of the ecosystem, such as species, biotopes formed byabitat-forming species or abiotic biotopes with a clear linkage toertain species.
The method to calculate an impact index value (I) for the set ofnthropogenic pressures in a given area was based on the followingormula (Halpern et al., 2008):
=n∑
i=1
m∑
j=1
Pi × Ej × �i,j
here Pi is the log-transformed and normalized value (scaledetween 0 and 1, and with 1 being the highest value of the pres-ure measured) of an anthropogenic pressure in an assessment unit, Ej is the presence or absence of an ecosystem component j (i.e.opulations, species, biotopes or biotope complexes; 1 or 0, respec-ively), and �i,j is the weight score for Pi in Ej (range 0–4, cf. Halpernt al., 2007). The impact of any P × E × � combination will be zero if
pressure is zero or an ecosystem component is absent. Thus, theore ecosystem components an area contains and the higher is the
umber of pressures in that area, the higher the index value. Thenal index value was calculated for 5 km × 5 km squares, i.e. thessessment units. A schematic presentation of the different stepsn the use of the index is given in Fig. 2.
.3. Compilation of data on anthropogenic pressures
We used the pressures to the marine environment as deter-ined in the MSFD (Anon, 2008); the directive recognizes 18
ressure types. The data on marine litter and ‘other substances’
Fig. 2. A schematic presentation of the different steps in the calculation of the indextool and its adaptations and suggested management purposes.
being unavailable and the impacts of non-indigenous speciesunknown, this study employed 15 pressure types (Table 1) dividedinto 52 Baltic Sea-wide data sets of anthropogenic pressures,which were direct pressure data, proxies for pressures, or merepresence/absence data of an activity or pressure. Baltic Sea envi-ronmental experts considered the data sets as evenly distributedand to include all relevant sources of each pressure. Since thereare no direct measurements for some of the pressures, they wereestimated on the basis of the causative human activities. Detailedinformation on the pressure data, including data sources, is pro-vided in Appendix A and in HELCOM (2010b).
2.4. Spatial distribution of species, biotopes and biotopecomplexes
There exists relatively accurate spatial distribution data onsome biotope complexes, biotopes and species in the Baltic Sea.Six benthic and two pelagic biotope complexes were chosen forthe index: photic sand, photic soft bottom, photic hard bottom,
non-photic sand, non-photic soft bottom, non-photic hard bot-tom, photic water column and non-photic water column (basedon Al-Hamdani and Reker (2007) and the EUSeaMap project). Twobenthic biotopes (mussel beds and Zostera meadows) and four
108 S. Korpinen et al. / Ecological Indicators 15 (2012) 105–114
Table 1Anthropogenic pressure types in the marine environment, included in this study.
Category Pressure type
Physical loss of seabed Smothering by dumped material, sealing of seabedPhysical damage to seabed Changes in siltation, abrasion of seabed, selective extraction of non-living resourcesOther physical disturbance Underwater noiseInterference with hydrological processes Changes in thermal regime, changes in salinity regimeContamination by hazardous substances Introduction of synthetic compounds, introduction of non-synthetic substances and compounds, introduction of
ganic
gens,
sp[hsiim
2
f2twedaa
FpAa
Lithuania, Poland, and Sweden). In addition, the scores were dis-cussed in an expert workshop, organized by HELCOM, and sixexperts at the HELCOM Secretariat gave a seventh set of scores.
0 20 40 60 80 10 0%
radio-nuclidesNutrient and organic matter enrichment Inputs of nutrients, inputs of orBiological disturbance Introduction of microbial patho
pecies-related distribution data sets (distribution of harbour por-oise [Phocoena phocoena], distribution of the three seal speciesgrey seal Halichoerus grypus, ringed seal Pusa hispida botnica andarbour seal Phoca vitulina], wintering grounds of sea birds andpawning and nursery areas of cod Gadus morhua) were includedn the index. Overall, there were 14 ecosystem components in thendex (Figs. 3 and 4, see details in Appendix B). All the distribution
aps have been published by HELCOM (2010b).
.5. Weighting coefficients
A weighting coefficient is a constant, which was used to trans-orm a pressure to a potential impact (Koskela, 2004; Halpern et al.,007; Vörösmarty et al., 2010). The weighting coefficient is specifico any combination of pressures and ecosystem components. Alleighting coefficients were based on a wide questionnaire among
xperts from all parts of the geographical area in question and from
ifferent fields of environmental sciences. The scores were given onscale from 0 to 4, reflecting no impact (0), low impact (1), moder-te impact (2), strong impact (3) or massive impact (4). When filling
ig. 3. Presence of ecosystem components (benthic and water column biotope com-lexes, benthic biotopes and species-related data layers) in 5 km × 5 km squares.ltogether 14 data layers were used (see text), but none of the squares containedll the ecosystem components.
matterselective extraction of species (e.g. fishing)
the questionnaire, experts gave consideration to three aspects insetting the score: recovery time of the ecosystem component afterthe pressure (<1, 1–5, 5–10 or >10 years), resilience of the ecosys-tem component to the pressure (very high, high, moderate, lowresilience or vulnerable) and the functional effect (i.e. whether thepressure affects one or several species, one or few trophic levelsor the whole community). An average of the three criteria was thefinal weighting coefficient.
The weighting coefficients were provided by national expertsin six countries around the Baltic Sea (Denmark, Estonia, Finland,
BOB
BOS
GOF
NBP
WGB
EGB
BOR
ARK
KAT
Zostera
Mussel
Pho�c sand
Pho�c so�
Pho�c hard
Apho�c sand
Apho�c so�
Apho�c hard
Cod
Seal
Porpoise
Bird
Fig. 4. Proportions of ecosystem components in the main sub- basins, defined astheir presence in all assessment units of the sub-basin. Since “photic water” coversentirely and “non-photic water” almost entirely all sub-basins, they were omittedfrom the graph. Sub-basins: Bothnian Bay (BOB), Bothnian Sea (BOS), Gulf of Finland(GOF), Northern Baltic Proper (NBP), Western Gotland Basin (WGB), Eastern GotlandBasin (EGB), Bornholm Basin (BOR), Arkona Basin (ARK) and the Kattegat (KAT). Forfull names of the ecosystem components, see Section 2.
al Indicators 15 (2012) 105–114 109
Ivct
2
tur9oiSbetbnraes2
3
3
btccnun
tiawBaiaoadop
eafiisTpezic
Fig. 5. Presentation of cumulative potential anthropogenic impacts by the Baltic Sea
S. Korpinen et al. / Ecologic
n order to include the greatest number of estimates, the medianalue of the expert estimates was chosen as the weighting coeffi-ient for each P × E combination. Methods and summary results ofhe questionnaire are presented in Appendix C.
.6. Data handling
The data on pressures and ecosystem components were linkedo a grid of the assessment units. Altogether 19,276 assessmentnits were calculated in the Baltic Sea area. Handling of the geo-eferenced data sets was done by the ESRI ArcGIS software, version.3, with the spatial analyst extension. In addition to the calculationf the index, spatial regression of the impact data layers (P × E × �n every cell) and the index result was performed by Ordinary Leastquares (OLS) method (see description and examples in ESRI, 2009)y analyzing data layers first one by one and then analyzing theffect of removing all data layers with R2 values less than 0.05 fromhe full model. R2 describes the percentage of variance explainedy the regression. In addition, the influence of ecosystem compo-ents and weighting coefficients were evaluated by two separateuns of Geographically Weighted Regression (GWR; see descriptionnd examples in Charlton and Fotheringham, 2009) with only onexplanatory variable in each model: the sum of normalized pres-ures or the full index with a constant weighting coefficient (value) instead of the ecosystem-specific coefficients.
. Results
.1. Spatial distribution of cumulative impacts
The cumulative impact values in the assessment units variedetween 6.8 (the Bothnian Bay) and 456.4 (the Sound) (Fig. 5),he theoretical maximum being 2912, assuming that all weightingoefficients and pressure values were maximal and all ecosystemomponents were present in an assessment unit. The minimumumber of anthropogenic pressures (value > 0) in an assessmentnit was 14 (in the offshore Bothnian Bay), whereas the maximumumber was 35 (in the Belt Sea, south of Zealand).
The spatial presentation of the cumulative impacts showed thathe highest potential impacts on the Baltic Sea ecosystem take placen the south-western sea areas (the Kattegat, Belt Sea, Kiel Baynd Mecklenburg Bay), the Gulf of Gdansk and the Gulf of Finland,hereas the least cumulative impacts were found from the Gulf ofothnia in the north (Fig. 5). The data indicated that the southernreas were under types of pressures which are rare or non-existentn the northern parts, such as bottom-trawling, large wind farmsnd large-scale extraction of seabed resources. Also other formsf commercial fishing were heavy in the southern sub-basins andtmospheric deposition of heavy metals and nitrogen occurred pre-ominantly in the southern areas. In the Gulf of Gdansk and Gulff Finland, the high sums of impacts were due mainly to riverineollution.
Another feature in the cumulative impact index was the differ-nce between the open sea and the coastal areas. In coastal areasll over the Baltic Sea, the multitude of coastal pressures – e.g.sh farms, municipal waste water treatment plants, river estuaries,
ndustries, warm-water outflows from power plants, and coastaltructures – create a heavy burden on the marine environment.he cumulative impacts were clearly higher in the coastal than theelagic areas in the Bothnian Bay, the Bothnian Sea and the North-
rn Baltic Proper. In other areas, cumulative impacts in the pelagicone were more or less as high as in the coastal areas. Moreover, thendex showed also high cumulative impacts in the vicinity of largeities, such as Copenhagen, Gdansk, St. Petersburg and Stockholm.Impact Index in 5 km × 5 km assessment units. The index in each assessment unitconsists of the sum of anthropogenic impacts on selected ecosystem componentspresent in the unit. See the index formula in Section 2.2.
The surface and midwater trawling (incl. long lines), bottomtrawling and shipping exerted pressures on the open-sea areas andlarge rivers such as the River Vistula, River Neva and River Götaseem to affect wide open-sea areas, as a result of pollution by nitro-gen, phosphorus and heavy metals (see Fig. 1 for rivers and coastalcities). In the Kattegat and Danish Straits the impacts had very highvariation among assessment units, reflecting numerous spatiallyrestricted activities like dredging, sand extraction and disposal ofdredged material (Fig. 6).
3.2. The effect of data determining the index result
The pressures which have the greatest contribution to the finalindex value are presented in Table 2. The sum of impacts of eachpressure type shows that the inputs of nutrients and organic matter,inputs of hazardous substances as well as fishing have the great-est overall impact on the Baltic Sea ecosystem (25%, 30% and 30%,respectively, Table 2). These pressures hence largely determine theoutcome of the map of potential cumulative impacts. According toR2, the highest-ranking data layers were those that exert pressureover the whole Baltic Sea (e.g. atmospheric deposition of metals),while the lowest ranks were given to the point impact data layers(e.g. dredging, construction sites and harbours). Including only 21impact data layers, which had R2 values over 0.05, resulted in theR2 value of 0.99 and showed the minor influence of point data forthe full index. Thus, the geographical distribution of a pressure ismore important in forming the shape of the map than its value ineach cell.
The ecosystem components have a central role in the indextool, because each added ecosystem component per assessmentunit introduces a full set of potential impacts to the index value.The GWR analyses showed that the pressure data layers without
110 S. Korpinen et al. / Ecological Indicators 15 (2012) 105–114
Table 2Summary statistics of the pressures. Minimum, maximum and mean impact scores and the standard deviation of them in the assessment units as well as the sum of impactsover the sea area are shown. The contribution of each of the pressures to the final index value is presented by coefficient of determination (R2). Each pressure has beenconverted to impacts in an assessment unit (see the index formula in Section 2). Pressures are ordered by the sum of impacts in all assessment units in the Baltic Sea.
Pressure type and pressure Min Max Mean SD Sum R2
Full model 6.81 456.44 111.52 48.86 2,149,710 1.00Extraction of species/mid- and surface water trawling + long lines 0 26.81 10.04 5.11 193,711 0.36Inputs of nutrients/atm. deposition of N 0 27.26 9.99 4.64 192,485 0.73Inputs of nutrients/waterborne N 0 25.54 8.78 4.23 169,190 0.65Inputs of org. Matter/Riverine organic matter 0 24.08 7.56 2.98 145,797 0.50Changes in siltation/organic matter 0 22.34 7.36 2.95 141,856 0.49Non-synthetic substances/atm. deposition of Pb 0 24.17 7.22 3.81 139,177 0.78Inputs of nutrients/waterborne P 0 24.79 6.83 3.12 131,663 0.48Extraction of species/gillnet fishery 0 25.14 6.15 4.68 118,574 0.46Extraction of species/bottom trawling 0 35.01 6.04 6.58 116,413 0.34Non-synthetic substances)/atm. deposition of Cd 0 18.37 5.82 3.05 110,533 0.77Non-synthetic substances/atm. deposition of Hg 0 21.86 5.06 2.96 96,187 0.62UW noise/all shipping 0 22.82 4.09 3.55 77,387 0.37Non-synthetic substances/waterborne Zn 0 20.38 4.00 2.97 75,914 0.11Non-synthetic substances/atm. depos. of dioxins 0 20.62 3.86 3.03 73,505 0.48Abrasion/bottom trawling 0 26.09 3.85 4.42 73,166 0.38Non-synthetic substances/waterborne Ni <0.01 19.65 3.19 2.97 60,585 0.05Extraction of species/Bird hunting 0 18.07 2.27 3.60 43,319 0.18Non-synthetic substances/waterborne Pb 0 19.71 2.17 2.86 41,274 0.14Extraction of species/trap and pot fishery 0 19.26 1.61 2.86 30,531 0.13Extraction of species/seal hunting 0 12.55 1.42 2.66 26,960 <0.01Non-synthetic substances/waterborne Cd 0 18.64 1.04 2.44 19,696 0.12Changes in siltation/coastal shipping 0 17.77 0.92 2.17 17,447 0.23Synthetic compounds/population density 0 17.51 0.36 1.49 6986 0.02UW noise/recreational boating 0 17.53 0.31 1.38 5941 0.03Introduct. of pathogens/passenger ships 0 15.84 0.31 1.01 5937 0.03Inputs of radioactive subst./radioactive substances 0 10.53 0.28 1.17 5363 0.03Non-synthetic substances/waterborne Hg 0 14.90 0.21 0.83 4084 0.05Changes in siltation/beaches 0 13.65 0.17 0.88 3242 0.04Synthetic compounds/harbours 0 25.40 0.17 1.33 3220 0.03Synthetic compounds/oil spills 0 16.13 0.16 0.88 3037 0.03Sealing/harbours 0 20.55 0.17 1.08 2577 0.50Sealing/coastal defence structures 0 24.77 0.13 1.33 2488 0.03Introduct. Of pathogens/Waste water treatm. plants 0 13.04 0.09 0.65 1638 0.02Changes in siltation/Dredging 0 20.22 0.08 0.93 1598 0.03Extraction of non-living resources/dredging 0 19.11 0.08 0.87 1482 0.03Abrasion/dredging 0 18.81 0.08 0.87 1478 0.03Changes in salinity/waste water treatm. plants 0 13.31 0.08 0.58 1439 0.03Smothering/Disposal of dredged matter 0 19.02 0.04 0.64 812 0.01Changes in salinity/coastal dams 0 20.00 0.04 0.56 674 <0.01Sealing/coastal dams 0 17.80 0.03 0.47 553 <0.01Inputs of nutrients/aquaculture 0 22.50 0.03 0.49 523 <0.01Inputs of org. matter/Aquaculture 0 22.50 0.03 0.48 512 <0.01Introduct. of pathogens/Aquaculture 0 8.35 0.02 0.28 399 <0.01UW noise/construction of cables 0 25.25 0.02 0.54 389 <0.01Smothering/construction of cables 0 24.20 0.02 0.44 314 <0.01Synthetic compounds/Industries 0 16.72 0.02 0.33 313 <0.01Changes in thermal regime/Nuclear power plants 0 18.20 0.01 0.27 97 <0.01UW noise/operational windfarms 0 12.20 <0.01 0.20 86 <0.01Synthetic compounds/polluting ship accidents 0 16.64 <0.01 0.20 69 <0.01UW noise/oil rigs 0 11.35 <0.01 0.14 37 <0.01
ettstrflF
wo2ai
Smothering/construction of windfarms 0
UW noise/construction of windfarms 0
cosystem components or weighting coefficients explained 48% ofhe total variation in the impact index (Figs. 7 and 8). The rest ofhe variation is caused by the ecosystem data and the ecosystem-pecific weighting coefficients. The cartographic presentation ofhe sum of normalized pressures (Fig. 3) shows a roughly similaresult as the results in the full impact index (Fig. 7). However, dif-erences are visible in areas where there were only very few or aarge number of ecosystem components present in the model (cf.ig. 3).
The weighting coefficients were based on expert scoring andere, thus, the only non-objective part of the index, their effect
n the index result was estimated by further testing (see Section.6). When adding a constant coefficient (value 2) to the indexnd comparing this dummy index with the real index, the dummyndex explained 99% of the real index. The test showed that the
14.50 <0.01 0.11 22 <0.0113.60 <0.01 0.11 20 <0.01
addition of the ecosystem components – without the variance fromthe weighting coefficients – explained more of the real index thanthe pressures alone. The remaining variance (1%) was explainedby the weighting coefficients and their effect on the impact mapwas only on details at the local level. The test results are shown inAppendix D.
4. Discussion
This study presented an assessment of cumulative potentialanthropogenic impacts in the Baltic Sea by an assessment tool
which uses three kinds of data on assessment units: anthropogenicpressures, ecosystem components and weighting coefficients totransform the pressures to impacts on each of the ecosystem com-ponents. The semi-enclosed nature of the Baltic Sea with its highly
S. Korpinen et al. / Ecological Indicators 15 (2012) 105–114 111
0 20 40 60
BOB
BOS
GOF
NBP
EGB
BOR
ARK
KAT
Mean index value per assessment unit
Fishing
Nutrient inputs
Non-synth e�cs
Synthe� cs
Physical damage
Physical loss
Organic ma�er
Hun�ng
Uw noise
Fig. 6. Mean index values of main pressure types per assessment unit in the BalticSea sub-basins. The pressures have been grouped to show the impact of varioushuman activities (see Tables 1 and 2, only the top nine pressure types are included).See the sub-basin names in Fig. 4.
Fig. 7. Presentation of the sum of normalized anthropogenic pressures. Each ofthe 52 pressures is on the scale from 0 to 1. The assessment units are 5 km × 5 kmsquares.
0 1 2 3 4 5
BOB
BOS
GOF
NBP
EGB
BOR
ARK
KAT
Mean of the total pressure
Fishing
Nutrient inputs
Non-synth e�cs
Synthe� cs
Physical damage
Physical loss
Organic ma�er
Hun�ng
Uw noise
Fig. 8. Total anthropogenic pressure in the Baltic Sea sub-basins. The normalized[scale: 0–1] pressure values have been first summed within an assessment unit
and then averaged per sub-basin. The pressures have been grouped according toTables 1 and 2; only the top nine pressure types are included. See sub-basin namesin Fig 4.developed and industrialised catchment area was well reflectedin this assessment of cumulative impacts: the highest cumulativeimpacts on the marine ecosystem were reported from the southernand south-western sea areas, with the highest population den-sities of up to 500 inhabitants per km2 in the drainage area. Incomparison, the drainages of the northern areas with the lowestcumulative impacts are very scarcely populated (2.0 inhabitants perkm2). Although the coastal areas always showed high cumulativeimpacts, the offshore areas were in a number of sub-basins equallyimpacted, reflecting the high fishing pressure, intensive maritimetraffic and large inputs of nutrients and hazardous substances fromthe atmosphere and rivers.
4.1. Reported ecosystem impacts of the dominant pressures
The Baltic Sea marine environment has been subject to scientificstudies for decades, but this is the first assessment of cumula-tive pressures and potential impacts in the region. Evaluation ofthe reliability of the tool requires comparison of the results withexisting knowledge on the state of the ecosystem. In this study,nutrient inputs were estimated to result in one of the highest pres-sures on the entire sea area. Recent studies have shown that theBaltic Sea suffers from a severe eutrophication problem (HELCOM,2009a; Lundberg et al., 2009; Andersen et al., 2010). According to a
recent assessment of the eutrophication status by the Baltic MarineEnvironment Protection Commission (HELCOM, 2009a), 176 of the189 studied sites were in an impaired state. The anthropogenicinputs of nitrogen and phosphorus are high in the entire sea area as
1 al Indi
nm(
BwefilicwlCrr
ic(oNiac(bt2at1
mBhea2her
4
msotmSa
jmiwtegsadta
12 S. Korpinen et al. / Ecologic
atural background inputs of nitrogen and phosphorus are esti-ated at only 17% and 16%, respectively, of the total inputs
HELCOM, 2004).According to our results, fishing exerts a high pressure on the
altic Sea ecosystem in all areas of the Baltic Sea. In this studye used fishing data from the year 2007, and at that time both
astern and western cod (Gadus morhua) stocks and three of theve assessed herring stocks (Clupea harengus) were below safe bio-
ogical limits due to overfishing and climatically induced changesn salinity and oxygen concentrations (ICES, 2007). The decline inod stocks has led to an increase in sprat (Sprattus sprattus) stocks,hich has been interpreted as a regime shift from a top-down regu-
ated food web to a resource-regulated one (Österblom et al., 2007;asini et al., 2008). The reproductive capacity of salmon spawningivers was mostly (84% of rivers) below the adequate level of rivereproductive capacity (75%) (ICES, 2007).
Bottom trawling was estimated to be an activity with a highmpact on the ecosystem. This was seen in the weighting coeffi-ient estimations by experts (Appendix C) and in the index resultsTable 2, Figs. 7 and 8). The result is in line with assessments fromther sea areas globally (Jones, 1992; Collie et al., 1997; Watling andorse, 1998; Jennings et al., 2001). In the Baltic Sea, bottom trawl-
ng by the Nephros fishery in the Kattegat and Skagerrak region canmount to a by-catch which is 50% of the biomass of the Nephrosatch and can include up to 24 non-target species in one catchOttosson, 2008). Harbour porpoise, grey seal, ringed seal, har-our seal and seabirds have all been found drowned throughouthe Baltic Sea in drift nets, gillnets and trawls (Lunneryd et al.,004; ICES, 2008; ASCOBANS-HELCOM database, 2011). Abrasionnd resuspension by bottom-trawling have been estimated as par-icularly destructive in the Baltic Sea (Riemann and Hoffmann,991; Tjensvoll et al., 2009) and globally (Watling and Norse, 1998).
Inputs of hazardous substances, particularly metals, were esti-ated to be among the highest pressures in all sub-basins of the
altic Sea. Adverse impacts of persistent organic pollutants andeavy metals have been found at all levels of the food web. Biomark-rs on molecular, cellular and tissue level have shown reproductivend other disorders in marine invertebrates and fish (Strand et al.,004; Broeg and Lehtonen, 2006), contaminant levels in biota areigh and populations of predatory species have only recently recov-red from reduced reproductive success (Helander et al., 2002,eviewed in HELCOM, 2010a).
.2. The role of weighting coefficients in the index
The weighting of pressures had two overarching aims: to esti-ate the degree of destructiveness of pressures and to specify the
ensitivity of ecosystem components to these pressures. As mostf these ecosystem components were described by abiotic factors,he weighting coefficients were estimated on the basis of the com-
unity or key species, which are typical to such a biotope complex.ubstrate type and light availability are key factors for benthic faunand flora (e.g. Al-Hamdani and Reker, 2007).
Estimations of weighting coefficients can be sensitive to sub-ectivity, as experts may judge impacts of pressures by different
eans. In order to avoid false scorings the expert questionnairencluded examples and several of the experts participated in a
orkshop. It is however obvious that a certain degree of subjec-ivity remains in the weighting coefficients. The analysis of theffect of the three factors in the index showed that the pressuresive the major shape for the index, the ecosystem componentsharpen the shape and the weighting coefficients only added detail
t a local scale. Thus, the non-objective component of the indexid not have a ruling effect on the final impact index result onhe scale of an entire sea area. It is however obvious that in a searea with strong environmental gradients, like the Baltic, regionalcators 15 (2012) 105–114
variability in impacts may be significant. Therefore regionallyadjusted weighting coefficients might improve the accuracy of themodel. The effects of the weighting coefficients are further pre-sented and discussed in Appendix D.
4.3. Strengths and weaknesses of the method
The method we used is based on a linear response ofecosystem components to anthropogenic pressures. This is a sim-plification of reality where thresholds or other non-linearitiescan be expected to occur. We think that these simplifica-tions do not however affect the main message of the endresult, which shows a clear geographical gradient of potentialcumulative impacts in the Baltic Sea. And overall, our studyemphasizes the need for further study of pressure-impact relation-ships and mechanisms, many of which have not been evaluatedscientifically.
The selection of the ecosystem components in the index affectsthe index result. Therefore the selection of ecosystem data hasbeen balanced among the different types of ecosystem compo-nents. By choosing only biotopes restricted to either coastal orpelagic areas or data on only few species one may bias the indexresults to certain geographical areas or parts of the ecosystem.We have avoided that by selecting benthic and water columnbiotope complexes which can be found both in the coastal and off-shore areas. Our benthic biotopes (i.e. Zostera meadows and bluemussel beds) were mostly from the coastal areas, but that wasbalanced by including pelagic species-related data layers. Thus,the high dependence of this index method on georeferenced data,may – in data-poor areas – lead to poor results. In the Baltic Sea,georeferenced data was, however, rather easily accessible due towell established regional cooperation among the coastal countries.In this study, the coarseness of some data layers was compen-sated by using a rather robust grid of 5 km × 5 km assessmentunits.
The method has strengths which should be considered againstthe weaknesses. There exists currently no other method whichgives an estimate of cumulative anthropogenic impacts and takesaccount of ecosystem sensitivity. The recent global assessment(Halpern et al., 2008) showed that the Baltic Sea was moderatelyaffected by the anthropogenic pressures, but being a global assess-ment it lacked regional data and therefore failed to show finer scalevariation within the region. By including Baltic Sea-specific pres-sure and ecosystem data we have shown that the tool can be usedto describe detailed differences at least at a basin-wide scale. Thetool has also been used for assessing other areas, such as in theEastern Pacific to map cumulative impacts in the California Current(Halpern et al., 2009) and Canadian waters (Ban et al., 2010) and toassess anthropogenic impacts on coral reefs in Hawaii (Selkoe et al.,2009).
The cumulative impact index is a tool which can be appliedto marine spatial planning, environmental impact assessments,placement and management of marine protected areas, permit-ting processes and also in environmental status, pressure andimpact assessments, such as those required by the EU MSFD.The index can be used to evaluate each ecosystem componentseparately, for example, by evaluating impacts of anthropogenicpressures on blue mussel beds. Often such practical uses requirea special selection of data layers depending on the aim of thepursuit. At smaller scales, the weighting coefficients may alsorequire adjustment to local conditions. The method can also beused to show potential impacts with predicted pressure data;
for example with climate change or ocean acidification scenar-ios. Together with spatial models capable of predicting spatialdrifting or extension of pressures, the index tool is a powerfulmethod.
al Indi
5
mtpeatdptMoDepe
A
Dit2UsMftc
A
t
R
A
A
A
A
B
B
C
C
C
C
C
S. Korpinen et al. / Ecologic
. Conclusions
The tool to assess cumulative anthropogenic impacts on thearine environment shows that the entire Baltic Sea ecosys-
em is under the burden caused by anthropogenic pressuresotentially resulting in high cumulative impacts on the wholecosystem. The approach which was first developed for globalssessments was here adapted to a regional sea, and the adapta-ion required modifications in data resolution, terminology andata management. The use of this tool provides a wide array ofossibilities for management and planning purposes from localo regional scales, thus offering also possibilities for the EU
ember States to estimate the level of anthropogenic impactsn their marine waters under the Marine Strategy Frameworkirective. Future assessments of cumulative impacts would ben-fit from a more detailed and quantitative understanding of theressure-impact relationships and mechanisms in the marinenvironment.
cknowledgements
We want to express our gratitude to the National experts inenmark, Estonia, Finland, Lithuania, Poland and Sweden for tak-
ng part in estimating the weighting coefficients and attendinghe HELCOM Baltic Sea Pressure Index Workshop on 11 February010 in Stockholm, Sweden. Thanks are also addressed to Ms.lla Li Zweifel (Dimedia) for the help in coordinating the expert
urvey and to Mr. Martin Isæus (AquaBiota Water Research),r. Henrik Skov (DHI) and Mr. Johnny Reker (Danish Agency
or Spatial and Environmental Planning) for some of the ecosys-em data layers. Special thanks to Ciarán Murray for linguisticorrections.
ppendix A. Supplementary data
Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.ecolind.2011.09.023.
eferences
l-Hamdani, Z., Reker, J. (eds.) 2007. Towards marine landscapes in the Baltic Sea.BALANCE Interim Report No 10. Available at: www.balance-eu.org.
ndersen, J.H., Axe, P., Backer, H., Carstensen, J., Claussen, U., Fleming-Lehtinen,V., Järvinen, M., Kaartokallio, H., Knuuttila, S., Korpinen, S., Laamanen, M.,Lysiak-Pastuszak, E., Martin, G., Møhlenberg, F., Murray, C., Nausch, G., Norkko,A., Villnäs, A., 2010. Getting the measure of eutrophication in the BalticSea: towards improved assessment principles and methods. Biogeochemistry,doi:10.1007/s10533-010-9508-4.
non, 2008. Directive 2008/56/EC of the European Parliament and the Council of 17June 2008 establishing a framework for community action in the field of marineenvironmental policy (Marine Strategy Framework Directive). Official Journal ofthe European Union, Brussels, L 164/19, 25.06.2008.
SCOBANS-HELCOM Harbour Porpoise Database and Map Service 2011. Availableat: http://www.helcom.fi/GIS/en GB/porpoise/.
an, N.C., Alidina, H.M., Ardron, J.A., 2010. Cumulative impact mapping: advances,relevance and limitations to marine management and conservation, usingCanada’s Pacific waters as a case study. Marine Policy 34, 876–886.
roeg, K., Lehtonen, K.K., 2006. Indices for the assessment of environmental pollutionof the Baltic Sea coasts: integrated assessment of a multi-biomarker approach.Marine Pollution Bulletin 53, 508–522.
asini, M., Lövgren, J., Hjelm, J., Cardinale, M., Moinero, J.-C., Kornilovs, G., 2008.Multi-level trophic cascades in a heavily exploited open marine ecosystem.Proceedings of the Royal Society B 275, 1793–1801.
harlton, M., Fotheringham, A.S., 2009. Geographically Weighted Regression: WhitePaper. National Centre for Geocomputation, National University of IrelandMaynooth, Available at http://ncg.nuim.ie/ncg/GWR/GWR WhitePaper.pdf.
oll, M., Lotze, H.K., Romanuk, T.N., 2008. Structural degradation in Mediterraneansea food webs: testing ecological hypotheses using stochastic and mass-balance
modelling. Ecosystems 11, 939–960.ollie, J.S., Escanero, G.A., Valentine, P.C., 1997. Effects of bottom fishing on the ben-thic megafauna of Georges Bank. Marine Ecology Progress Series 155, 159–172.
rain, C.M., Halpern, C.S., Beck, M.W., Kappel, C.V., 2009. Understanding and manag-ing human threats to the coastal marine environment. The Year in Ecology and
cators 15 (2012) 105–114 113
Conservation Biology, 2009. Annals of the New York Academy of Sciences 1162,39–62.
ESRI. 2009. Ordinary Least Squares (Spatial Statistics). ArcGIS Desk-top 9.3 Help. Last modified in January 28, 2009. Available at:http://webhelp.esri.com/arcgisdesktop/9.3/.
Halpern, B.S., Kappel, C.V., Selkoe, K.A., Micheli, F., Ebert, C.M., Kontgis, C., Crain, C.M.,Martone, R.G., Shearer, C., Teck, S.J., 2009. Mapping cumulative human impactsto California Current marine ecosystems. Conservation Letters 2, 138–148.
Halpern, B.S., Selkoe, K.A., Micheli, F., Kappel, C.V., 2007. Evaluating and ranking thevulnerability of global marine ecosystems to anthropogenic threats. Conserva-tion Biology 21, 1301–1315.
Halpern, B.S., Walbridge, S., Selkoe, K.A., Kappel, C.V., Micheli, F., D’Agrosa, C., Brunoa,J.F., et al., 2008. A global map of human impact on marine ecosystems. Science319, 948–952.
Helander, B., Olsson, A., Bignert, A., Asplund, L., Litzén, K., 2002. The roleof DDE, PCB, coplanar PCB and eggshell parameters for reproductionin the white-tailed sea eagle (Haliaeetusalbicilla) in Sweden. Ambio 31,386–403.
HELCOM. 2004. The fourth Baltic Sea pollution compilation (PLC-4). Baltic SeaEnvironment Proceedings No. 93, 189 pp. Helsinki Commission. Available at:www.helcom.fi/publications.
HELCOM. 2009a. Eutrophication in the Baltic Sea – An integrated thematicassessment of the effects of nutrient enrichment in the Baltic Sea region.Baltic Sea Environment Proceedings 115B.Helsinki Commission. Available at:www.helcom.fi/publications.
HELCOM. 2009b. Biodiversity in the Baltic Sea – An integrated thematicassessment on biodiversity and nature conservation in the Baltic Sea.Baltic Sea Environment Proceedings 116B.Helsinki Commission. Available at:www.helcom.fi/publications.
HELCOM. 2010a. Hazardous substances in the Baltic Sea – An integratedthematic assessment of hazardous substances in the Baltic Sea. BalticSea Environment Proceedings 120B. Helsinki Commission. Available at:www.helcom.fi/publications.
HELCOM. 2010b. Towards a tool for quantifying the anthropogenic pressures andpotential impacts in the Baltic Sea marine environment. A background docu-ment on the method, data preparation and testing of the Baltic Sea Pressureand Impact Indices. Baltic Sea Environment Proceedings 125. Available at:www.helcom.fi/publications.
ICES, International Council for the Exploration of the Sea. 2007. ICES Advice 2007,Book 8. Available at: www.ices.dk.
ICES, International Council for the Exploration of the Sea. 2008. Interactions betweenfisheries and seabirds in EU waters. ICES Advice 2008, Book 1. Available at:www.ices.dk.
Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque,B.J., Bradbury, R.H., Cook, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S.,Lange, C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Stenbeck, R.S., Tegner,M.J., Warner, R.R., 2001. Historical overfishing and the recent collapse of coastalecosystems. Science 293, 629–637.
Jennings, S., Pinnegar, J.K., Polunin, N.V.C., Warr, K.J., 2001. Impacts of trawling dis-turbance on the trophic structure of benthic invertebrate communities. MarineEcology Progress Series 213, 127–142.
Jones, J.B., 1992. Environmental impact of trawling on the seabed: a review. NewZealand Journal of Marine and Freshwater Research 26, 59–67.
Kappel, C.V., 2005. Losing pieces of the puzzle: threats to marine, estuar-ine, and diadromous species. Frontiers in Ecology and the Environment 3,275–282.
Koskela S. (ed.), 2004. Environmental analysis and indicators for the Kymenlaaksoregion. Documentation report 1 of the ECOREG project. The Finnish Environ-ment, 697, 149, Available at www.ymparisto.fi.
Lindeboom, H.J., 2005. Comparison of effects of fishing with effects of natural eventsand non-fishing: anthropogenic impacts on benthic habitats. American FisheriesSociety Symposium 41, 609–617.
Lundberg, C., Jacobsson, B.-M., Bonsdorff, E., 2009. The spreading of eutrophicationin the eastern coast of the Gulf of Bothnia, northern Baltic Sea – An analysis intime and space. Estuarine, Coastal and Shelf Science 82, 152–160.
Lunneryd, S.G., Königson, S., Sjöberg, N. 2004. Bifångst av säl, tumlare och fåglar idet svenska yrkesfisket. (Bycatch of seals, porpoise and sea birds in the SwedishFishery.) In Swedish with an English summary. Finfo. Fiskeriverket informerar8:1:21. Available at: http://www.fiskeriverket.se.
Millenium Ecosystem Assessment. 2005. Ecosystems and human well-being: Syn-thesis report. World Resources Institute. Washington, DC.
Österblom, H., Hansson, S., Larsson, U., Hjerne, O., Wulff, F., Elmgren, R., Folke, C.,2007. Human-induced trophic cascades and ecological regime shifts in the BalticSea. Ecosystems 10, 877–889.
Ottosson L., 2008. By-catches of non-commercial invertebrate taxa in Skagerrak andKattegat, generated by demersal otter trawling. Master Thesis in marine ecology,Department of Marine Ecology, Gothenburg University, Sweden, 67 p.
Riemann, B., Hoffmann, E., 1991. Ecological consequences of dredging and bot-tom trawling in the Limfjord, Denmark. Marine Ecology Progress Series 69,171–178.
Selkoe, K.A., Halpern, B.S., Ebert, C.M., Franklin, E.C., Selig, E.R., Casey, K.S., Bruno, J.,
Toonen, R.J., 2009. A map of human impacts to a pristine coral reef ecosystem,the Papahanaumokuakea Marine National Monument. Coral Reefs 28, 635–650.Strand, J., Andersen, L., Dahllöf, I., Korsgaard, B., 2004. Impaired larval developmentin broods of eelpout (Zoarces viviparus) in Danish coastal waters. Fish Physiologyand Biochemistry 30, 37–46.
1 al Indi
T
V
Global threats to human water security and river biodiversity. Nature 467,
14 S. Korpinen et al. / Ecologic
jensvoll, I., Bradshaw, C., Sköld, M., Molvaer, J., Nilsson, H.C., 2009. Sediment resus-
pension by trawling: importance for contaminants and nutrient remobilisationin the Baltic Sea. In: Abstract to the 7th Baltic Sea Science Conference, 17–21August 2009, Tallinn, Estonia, Abstract available at http://www.bssc2009.org/.örösmarty, C.J., McIntyre, P.B., Gessner, M.O., Dudgeon, D., Prusevich, A., Green,P., Glidden, S., Bunn, S.E., Sullivan, C.A., Reidy Liermann, C., Davies, P.M., 2010.
cators 15 (2012) 105–114
555–561.Watling, L., Norse, E.A., 1998. Disturbance of the seabed by mobile fish-
ing gear: a comparison to forest clearcutting. Conservation Biology 12,1180–1197.
Chapter 15
Ecosystem health Maria Laamanen, Samuli Korpinen, Ulla Li Zweifel and Jesper H. Andersen
Chapter summary Humans have inhabited the Baltic Sea drainage area for several thousand years but it has not been until in recent times that the impacts from human activities have surpassed the natural dynamics of the ecosystem. Human induced degradation of the ecosystem health accelerated after the World War II. According to overview assessments of ecosystem state in the 2000s, human pressures and impacts can be found in all areas of the Baltic Sea. Due to the human pressures most of the Baltic Sea area is a eutrophication problem area, it is disturbed by hazardous substances, and has impoverished biodiversity status and an impaired overall ecosystem health. The use of the Baltic Sea ecosystem has not been at a sustainable level. This is evidenced also by the ecosystem regime shifts in the late 20th century caused by fishing, hunting and eutrophication, as well as changes in climatic conditions. Although some of the ecosystem impacts from human activities such as local sewage pollution problems, contamination by DDT and polychlorinated biphenyls (PCBs) have been salvaged, the widespread eutrophication with bottom hypoxia and shifts in biodiversity are likely to pose much greater challenges. There is currently international legislation and regional cooperation in force that has explicit aims and measures for a healthier marine ecosystem.
15.1 Ecosystems are naturally dynamic
15.1.1 Natural fluctuations occur at various time scales Ecosystem functions and processes are dynamic by nature. Time scales of ecosystem processes can be geological, centennial, decennial, annual, seasonal or daily depending on the process in question. The Baltic Sea has a long history of interchanging marine and freshwater phases and the salinity levels and climate conditions close to the current ones have existed only for about 3 000 years. The changes between marine and freshwater phases were driven by natural changes in the morphology and depth of the Baltic Sea basin which regulated the connectedness of the basin to the ocean and the levels of salinity. Four geological phases of the Baltic Sea have been distinguished since the recent glacial each with their distinct fauna and flora. The Baltic Sea today is a young ecosystem and it has a relatively simple ecosystem structure and few species compared to some of the stabilised old marine ecosystems such as tropical sea areas with coral reefs or temperate seas with kelp forests. The northernmost areas of the Baltic Sea are still subject to land uplift after having been depressed by the masses of ice for several thousand years during the recent glacial that
1
ended about 11,000 years ago in the Baltic region. In the north-eastern Bothnian Bay coast, the rising of the Earth’s crust can be 9 mm per year while in some areas of the southern rim of the Baltic Sea coast, this change is opposite and land is sinking approximately at a rate of 1 mm/year (Schmidt-Thomé 2006). Thus, the Baltic Sea coast line is subject to a constant directional change. During the last one hundred years the salinity, oxygen and temperature conditions of the Baltic Sea have varied on a decadal time-scale (Winsor et al. 2001). These changes in the hydrography in turn cause changes in the abundance and distribution of both pelagic, benthic and littoral species and communities in the Baltic Sea (Alheit et al. 2005). The climatically driven inflows of saline water from the North Sea are crucial for ecosystem structure and functions and they act through changes of the physical and chemical conditions of the water body. The saltwater pulses flow into the Baltic along its bottoms, from one deep sub-basin into another, slowly mixing with the brackish water and oxygenising the deep waters and sediment. Increasing salinity of the deep water also strengthens the vertical physical structure of the water column and the physical barrier across the halocline, diminishing vertical mixing in the water column, and in the longer run, making the bottom water vulnerable to hypoxia (< 2 mg l-1 oxygen) and complete anoxia. Hypoxic conditions are physiologically intolerable to most fish and invertebrate species and are only inhabited by procaryotes such as sulphur bacteria. The spatial extent of anoxic bottoms has varied over time and changes in the morphology and depths of the Baltic basin and the sills have been the main triggers for millennial scale changes in hypoxia (Zillén & Conley 2010). On a decadal scale, periods without saline water inflows, so called stagnation periods, have been observed in the Baltic from the late 1970s to the early 1990s. In the Baltic Proper, the lower frequency of saline water inflows has tended to result in freshening of both the surface and bottom waters, weakening and deepening the positioning of the halocline, and most importantly relieving the hypoxia of the waters lying beneath the halocline (Conley et al. 2002). Fluctuations over the seasonal scale are also pronounced in the Baltic Sea. They are mainly driven by changes in temperature and to some extent by changes in precipitation. These changes cause seasonality in the physical structuring of the water column. Probably the most pronounced seasonal phenomenon is the formation of a thermocline in spring which evokes the spring bloom of planktonic algae and initiates the period of higher productivity in the pelagic zone. The spring bloom is an important phenomenon because it serves the benthic communities with energy by sedimentation of the dead organic matter after the bloom. The thermocline breaks down in the autumn and is followed by a full turnover and mixing of the water column down to the halocline level. In the wintertime parts of the Baltic are covered by ice each year but the extent of ice varies greatly from one year to another. There are also functions that fluctuate on a diel scale, such as vertical migration of zooplankton or primary production of phytoplankton or macroalgae. These diel variations are largely driven by changes in solar radiation and resulting light conditions underwater.
2
15.2.1 Ecosystem dynamics have various spatial scales There are differences in the spatial scales of various natural phenomena and the scale is related to the drivers of these processes. Salt water inflows from the North Sea tend to have whole Baltic Sea scale influence although their impacts in the northernmost sub-basins of the Gulf of Bothnia are less pronounced and take place only after a time lag. Seasonal changes influence the whole Baltic Sea but there are spatial differences in the way these dynamics are expressed. The growth season is the longest in the south and shortest in the north which is manifested by an earlier induction of the thermocline and onset of spring blooms in the south than in the north as well as the later timing of the autumn turnover. At a local scale, a sheltered shallow bay is likely to undergo spring phenomena earlier than the open sea environment nearby because solar radiation will heat the shallow less turbulent water of the bay faster than waters of the deeper, more wind exposed open sea nearby and the bay may also be subject to inflowing fresh river waters that will further strengthen the physical layering of its water column. Big inflowing rivers such as the rivers Neva, Oder or Vistula tend to have sub-basin scale influence. Changes or perturbations in the drainage area, such as flooding, often influence at the sub-basin scale through the discharge of river waters. The coastal zone tends to host ecological phenomena that function at smaller spatial scales than those of the open sea. This is largely due to the mosaic-like geographical structure of the coastal zone, providing to the biota a physically much more varied environment than that of the open sea environment. As a result, the habitat complexity as well as species diversity tends to be greater in the coastal zone than in the open sea.
15.2 Human pressures on the marine ecosystem
have a long history
15.2.1 Human settlement and evolving human activities The human settlement in the Baltic Sea catchment developed as humans followed the rim of the diminishing glacial ice shelf towards the north and inhabited the Baltic Sea basin approximately 9,000 to 13,000 years ago. A population expansion took place in the Baltic Sea drainage area between 1000 and 1300 AD increasing the human population from 4.6 to 9.5 million people. But it was not until between 1700 and present time that a great population boom took place increasing the population in the drainage basin from about 14.4 to 85 million (Zillén & Conley 2010). The sea has been a resource for coastal populations. Fisheries and hunting of seals and harbour porpoises as well as water birds has always been part of the culture of the coastal and archipelago communities. The increased number of humans in the catchment was naturally accompanied by an increase in the use of land for agriculture and forestry through land reclamation, and more recently by urbanization and industrialization. The human
3
activities on the catchment area have increased the inflow of soil particles, nutrients and more recently of pollutants to the sea. As long as there have been humans, the Baltic has also served as a transportation route and connected peoples living on its shores. Just like the other human activities, also shipping activity has intensified after the 1950s and today there are about 2000 ships plying the Baltic Sea at any time point. Shipping results in losses to the water of nutrient rich sewage, emissions to the air of nitrogen and sulphur, and it promotes the transfer to the Baltic Sea of alien species from other sea regions. In coastal areas and shallower off-shore areas it causes erosion and resuspension of bottom sediments. So far, catastrophic shipping accidents with large-scale ecosystem effects have been avoided but pollution by petrogenic substances in smaller scale has been common throughout the time of motorized ships. After the World War II, technological development intensified the agriculture, forestry, shipping, as well as fisheries to an extent unthinkable in the times when the use of the sea was more subsistence orientated. The exploitation of the sea became industrialised. All of this has also had drastic impacts on the marine ecosystem. Signs of deterioration were first seen near the cities due to the pollution by untreated waste waters. Contemporary inputs of total nitrogen (N) to the Baltic Sea have been estimated to be more than twice the amount a century ago and total phosphorus (P) about three times the quantity a century ago (Savchuk et al. 2008). Anthropogenic climate change puts a further pressure on the Baltic Sea by changing the physical setting of the sea and through that changing also the chemical and biological features of the ecosystem. The air temperature in the Baltic Sea region has warmed by 0.08°C per decade on average during the period 1871–2004 (Heino et al. 2008). This trend is slightly steeper than the trend in the global time series where there has been a warming of the global climate system by 0.74°C degrees over the period 1906–2005 (IPCC 2007). During recent decades, there has been a decreased frequency of salt-water pulses from the North Sea into the Baltic Sea. The length of the ice season has decreased by 14–44 days during the last century. These changes have had a measurable effect on the distribution, reproductive output and stock sizes of the Baltic biota. However, it has not been possible to establish a distinct causal link between these changes and anthropogenically induced climate changes, partly because of the large natural climate variability, but also owing to possible impacts from other human pressures (Dippner et al. 2008).
15.2.2 Baltic Sea regime shifts Ecosystems tend to exhibit stable ecosystem states but if heavily stressed by external factors they can undergo drastic changes that reorganise the ecosystem structure and functioning (e.g. Scheffer et al. 2001). The causative external factors can be natural, such as climatic fluctuations, or anthropogenic stressors, or combinations thereof. A healthy ecosystem has an inherent capacity to buffer and counteract disturbances. This capacity of an ecosystem to absorb change and to recover from it is called ecosystem resilience (e.g. Elliott et al 2007). Biological diversity and variability in an ecosystem tends to build up ecosystem resilience. As a consequence of resilience, ecosystems can appear virtually unaffected and stable while exposed to considerable stress. The apparent lack of
4
response can be explained by natural feedback mechanisms such as biogeochemical compensation, regulation through trophic and competitive interactions within the system, and, to a certain degree, also by the functional diversity and redundancy among species. Regime shifts take place when the capacity of the ecosystem to absorb and buffer external pressures is eventually exceeded, in other words when the resilience of the system has been surpassed. At a certain point even a small increment in external pressure can cause a shift that will result in the collapse or dramatic change of populations or other characteristics of the ecosystem. An idiom that reflects this type of event is “the straw that broke the camel’s back”. The stress level that pushes the system into a change is often referred to as a threshold or tipping point (Fig. 15.1B and D). Below the threshold level one stable state prevails and above the threshold another stable state. Once a new state has been reached the system has a tendency to be self-reinforcing as feed-back mechanisms start stabilising the new regime. Ecosystem changes can also be gradual with no apparent thresholds or tipping points (Fig. 15.1A and C). In the Baltic Sea, it has been suggested that several regime shifts have occurred during the past 80 years, likely driven by variability in climate, eutrophication, as well as seal hunting and fishing pressure (Österblom et al. 2007, ICES 2008). Two shifts were observed as late as in the end of the 1980s and the mid-1990s. In the central Baltic Sea, The period before the first shift was characterised by relatively high cod and herring spawner biomass and recruitment, and high abundances of the copepod Pseudocalanus acuspes. The period after the
Figure 15.1 Conceptual models of the possible modes
of responses of ecological status in regard to different
dose-response relationships. The dashed line indicates an
environmental target for ecosystem quality, presented as
an Ecological Quality Ratio, and the red arrows indicate
the estimated reductions in pressures needed to meet the
target. The reductions increase from scenario A to D
indicating that fulfilment of the target in non-linearly
responding systems with a threshold (scenario B) or a
shifting baseline (scenario D) or combinations hereof
(scenario D), e.g. caused by climate change, call for
reductions significantly larger compared to linearly
responding systems (scenario A). Figures are based on
Duarte et al. (2009) and Kemp et al. (2009) and the figure
has been created with the assistance of Janus Larsen,
NERI, Denmark.
5
second shift has seen a sprat dominance of the system with high abundances of Acartia spp. and Temora longicornis zooplankton species. The decrease in the populations of top predators like seals has modified the trophic structure of the Baltic Sea food webs. Examples from around the world show that decimation of top predators from the food webs through hunting and fishing have made the marine ecosystems vulnerable to other threats like pollution by hazardous substances, eutrophication and arrival of alien species (Jackson et al. 2001). At a larger scale, the ecosystem has changed over the past one hundred years from an oligotrophic clear water sea with abundant populations of top predators into a heavily eutrophicated sea area burdened by pollution of hazardous substances and with an increasing quantity of alien species.
15.3 Present anthropogenic pressures and state of
the ecosystem
15.3.1 Human pressures on the ecosystem Numerous pressures from various types of human activities acted on the Baltic Sea marine ecosystem in the 2000s (Table 1). These pressures tend to result in eutrophication, pollution effects and deterioration of biodiversity and some of the single pressures may have multiple and even synergistic effects. For example nutrient inputs cause eutrophication as well as changes in biodiversity through, e.g. overgrowth of certain opportunistic species. Overfishing, on the other hand, may exacerbate the effects of eutrophication by changing the food web structure. The impacts may be direct or indirect, acting through other effects. For example dredging has a direct physical impact on the benthic environment harming organisms living in the sediment. But dredging may also have an indirect effect of releasing contaminants buried in the sediments to the environment (Table 1). At present, commercial fishing, nutrient inputs, smothering, and pollution by hazardous substances are the most adverse pressures acting on the Baltic ecosystem according to expert views compiled by HELCOM (HELCOM 2010a, b). These views are in line with studies on top pressures from marine regions in the world. Fishing, nutrient inputs and inputs of hazardous substances are all intense activities with wide-spread impacts. In addition they have significant impacts on many different ecosystem components. The actual quantity of pressures on the ecosystem, such as smothering of the bottom habitats, is in many cases difficult to measure. Therefore, human activities that act as drivers of the pressures are measured instead, or other proxies are used. This approach was used to estimate the quantity and distribution of all potential pressures in the Baltic Sea, when HELCOM created a Baltic Sea Pressure Index, BSPI (HELCOM 2010b). In the Baltic Sea, the sum of pressures tends to be the greatest in the eastern and southern areas of the Baltic, especially in the Gulfs of Finland, Gdansk and Riga, as well as the Sound (Figure
6
Table 15.1 Human-derived pressures on the Baltic Sea marine ecosystem with their drivers (human activity)
and potential impacts on eutrophication (E), pollution effects by hazardous substances (HS) and biodiversity
(BD). x = direct impacts; (x) = indirect impacts. From HELCOM (2010a). Pressure Human activity, proxy or a direct measure of the
pressure Potential impacts
E HS BD Smothering Wind farms, bridges, oil platforms (construction phase) (x) x
Smothering Cables and pipelines (construction phase) (x) x
Smothering Disposal of dredged material x x
Sealing Coastal defense structures x
Sealing Harbours x
Sealing Bridges x
Changes in siltation Shipping (coastal) (x) x
Changes in siltation Riverine input of organic matter x
Changes in siltation Bathing sites, beaches and beach replenishment x
Changes in siltation Dredging, sand, gravel or boulder extraction (x) x
Abrasion Dredging, sand, gravel or boulder extraction (x) x
Abrasion Bottom trawling (x) (x) x
Selective extraction Dredging, sand, gravel or boulder extraction resulting in, e.g., habitat loss
x
Underwater noise Shipping (coastal and offshore) x
Underwater noise Recreational boating and sports x
Underwater noise Cables and pipelines (construction phase) x
Underwater noise Wind farms, bridges, oil platforms (construction phase) x
Underwater noise Wind farms (operational) x
Underwater noise Oil platforms x
Changes in thermal regime Power plants with warm-water outflow x
Changes in salinity regime Bridges and coastal dams x
Changes in salinity regime Coastal wastewater treatment plants with freshwateroutlets to the sea
x
Introduction of synthetic compounds Polluting ship accidents x x
Introduction of synthetic compounds Coastal industry, oil terminals, refineries, oil platforms x x
Introduction of synthetic compounds Harbours x x
Introduction of synthetic compounds Atmospheric deposition of dioxins x (x)
Introduction of synthetic compounds Population density (e.g., hormones and pharmaceuticals) x (x)
Introduction of non-synthetic substances and compounds
Illegal oil spills x x
Introduction of non-synthetic substances and compounds
Waterborne input of Cd, Hg and Pb x x
Introduction of non-synthetic substances and compounds
Atmospheric deposition of Cd, Hg and Pb x (x)
Introduction of radionuclides Discharges of radioactive substances x
Inputs of nutrients Waterborne input of nitrogen x x
Inputs of nutrients Waterborne input of phosphorus x x
Inputs of nutrients Aquaculture x x
Inputs of nutrients Atmospheric deposition of nitrogen x x
Inputs of organic matter Aquaculture x x
Inputs of organic matter Riverine input of organic matter x x
Introduction of microbial pathogens Coastal wastewater treatment plants with outlets to the sea x
Introduction of microbial pathogens Aquaculture x
Selective extraction of species Bottom trawling (landings or catches) (x) x
Selective extraction of species Surface- and mid-water trawling (x) x
Selective extraction of species Gillnet fishery (x) x
Selective extraction of species Coastal stationary gear fishery x
Selective extraction of species Hunting of seals x
Selective extraction of species Hunting of birds x
15.2A). The Bothnian Bay and Bothnian Sea as well as the northern Baltic Proper are the areas with the lowest pressures. These trends can largely be explained by the higher
7
densities of populations in the catchments in the south and east than in the north. Altogether 52 different pressures or activities were included in this analysis.
A
B
Figure 15.2 Sum of human pressures and their potential impacts in the Baltic Sea. Panel A: the Baltic Sea
Pressure Index (BSPI). Panel B: the sum of potential impacts from these pressures on the marine ecosystem
according to the Baltic Sea Impact Index (BSII), with smaller values (blue), indicating a lower and higher values
(red) a greater sum of pressures. From Korpinen et al. (2011).
Assessment of potential impacts on the ecosystem from these pressures requires knowledge about the distribution of ecosystem components in addition to knowledge of the distribution of the pressures. Even more important, it also requires knowledge or estimates of the severity of the impacts from each human pressure or activity on each ecosystem component. HELCOM derived these relationships, so called impact factors, through expert estimates while creating the Baltic Sea Pressure Index and the associated Baltic Sea Impact Index (cf. HELCOM 2010b). The Impact Index provides an estimate of the sum of potential impacts from human pressures on the marine ecosystem and offers a way to present the spatial distribution of these impacts. In the Baltic Sea, the coastal areas host a larger number of ecosystem components than the open sea and consequently these areas are also more vulnerable to human pressures (Figure 15.2B, HELCOM 2010a). Also many of the local scale pressures, such as dredging and construction of harbours or marinas takes place in the coastal zone in addition to larger scale pressures. In the open sea, most of the impacts are related to fishing or airborne deposition of nitrogen or hazardous substances.
15.3.2 Nutrient inputs and eutrophication Nutrient loading to the Baltic Sea peaked in the 1970s to 1980s and today’s loads are lower than those two decades ago (HELCOM 2011, 2012a). Between the years 1990 and 2006
8
phosphorus loads were estimated to have declined by 45% and nitrogen by 30% (HELCOM 2009a). In 2006, 638,000 tonnes of N and 28,400 tonnes of P still entered the Baltic Sea via waterways (HELCOM 2011). In addition, 200,000 tonnes of airborne N was deposited onto the Baltic Sea in 2006 (Bartnicki 2011). The fraction of P entering the Baltic Sea via atmospheric pathways is usually minor. These input levels are still far above the so called maximum allowable nutrient input levels of the HELCOM Baltic Sea Action Plan (BSAP, see section 15.5.4). According to the BSAP, 601,000 tonnes of N and 15,300 tonnes of P are estimated to safeguard a Baltic Sea with an acceptable level of eutrophication (HELCOM 2007a, Backer et al 2010). Point sources, mainly municipal and industrial waste waters used to be the main sources of phosphorus and a large source of nitrogen to the Baltic Sea but due to the improved waste water treatment during the recent decades agriculture has taken over the role of the single most polluting sector for both nutrients. In 2006, losses from diffuse sources were the main origin of the excessive inputs of both nutrients (at least 45 % for both nutrients) and only about 12 % of the nitrogen and only 20 % of phosphorus loads originated from point sources (HELCOM 2011). Agriculture was the major sector contributing on average 60-90% of the diffuse inputs of both nitrogen and phosphorus. In addition, scattered dwellings and storm waters are also significant sources. The main countries contributing to the waterborne nutrient input in 2006 were Poland (24% of N, 36 % of P), Sweden (19% of N, 13% of P), and Russia (17% of N, 14 % of P) (HELCOM 2011). Between 1994 and 2008 only Denmark showed significant decreasing riverine loads, while Estonia and Finland had significant increasing riverine loads. The greatest area-specific nutrient losses are found in catchment areas with high population density, many industries and high agricultural activity, such as the south-western part of the Baltic Sea catchment. Eutrophication means an increase in the rate of supply of organic material to the water (Nixon 1995). In the Baltic Sea, anthropogenic inputs of N and P have resulted in elevated levels of both nutrients in the water. Excess nutrients and the subsequent accelerated primary production cause the well-known eutrophication effects like increased biomasses of planktonic as well as periphytic algae, increased frequency and intensity of algal blooms, decreased water transparency, increased secondary production and upsurge in the biomass of e.g. cyprinid fishes. In the end of the chain of effects is hypoxia and often even complete anoxia in the bottom waters and sediments since decomposition by bacteria of the increased organic matter consumes the oxygen (HELCOM 2009a). It seems that the 1950s was the turning point in the rate of increase of nutrient loading and human-induced eutrophication of the Baltic Sea (HELCOM 2012a). Actual long term measurements going beyond the 1950s are few but acceleration of primary production is reflected in the organic carbon accumulation rate over the last 40 years in the sediment records from the Bornholm Basin and Gotland Deep (Struck et al. 2000). Increased eutrophication during the 20th century is also reflected in the water transparency where declines of summer Secchi depths have been observed in all sub-basins of the Baltic Sea over the last one hundred years (Fleming-Lehtinen et al. 2010). The decrease has been most
9
pronounced in the northern Baltic Proper (from 9 to 5 m) and the Gulf of Finland (from 8 to 4 m). Eutrophication is also reflected in the increase of the total hypoxic (less than 2 mg O2 l-1) area from less than 10,000 km2 in 1900-1910 to 35,000 km2 in the 1950s and to around 60 to 70,000 km2 in the recent years (HELCOM 2012a). In 2001-2006, most of the Baltic Sea area was affected by eutrophication (HELCOM 2009a). Only 13 of 189 areas assessed by HELCOM had an acceptable eutrophication status. Those areas were found in the open Bothnian Bay and the north-eastern Kattegat, as well as in coastal sites in the Gulf of Bothnia. The Gulf of Bothnia is an area with the least human pressures and the smallest population in the catchment area, while the western sea area, the Kattegat, is influenced by the Atlantic waters (HELCOM 2009a). The eutrophication state was evaluated through the use of data on indicators such as nutrient concentrations, summer water transparency, chlorophyll a levels, maximum depth penetration of macrophytes and species diversity of macrozoobenthic invertebrates in the deep bottoms or the use of various macrozoobenthos indices in the coastal zone (HELCOM 2009a). Each of the indicators had an associated target level indicating the border between an “acceptable” and “non-acceptable” eutrophication status. These target levels assisted in the integration of the indicator information into a eutrophication assessment for each assessment area (cf. Andersen et al. 2010). A particular challenge for remedial actions in the Baltic Sea is that eutrophication, through increased sedimentation of organic matter, has contributed to extension of hypoxic areas, as well as nutrient enrichment of the sediments (HELCOM 2009a, 2012a). When sediments become hypoxic, phosphate is released to the overlaying water mass, primarily through dissimilatory reduction of iron oxyhydroxides by bacteria (Conley et al. 2002). Phosphate thus becomes available for growth of primary producers that thrive in nitrogen deficient conditions. These kind of conditions in particular fuel the blooms of nitrogen fixing cyanobacteria which not only are self-sufficient in terms of nitrogen but also can increase nitrogen levels in the pelagic system. Thus, there are currently processes in place that act to maintain the new eutrophic regime (Vahtera et al. 2007). In order to revert to a less eutrophicated state it may necessary for the nutrient loads to be reduced to levels that are lower than those that existed before the regime shift (Figure 15.1D).
15.3.3 Pollution and effects by hazardous substances Chemical substances can be considered hazardous if they are toxic, persistent and bioaccumulate, or if they are very persistent and bioaccumulate. In addition, substances with effects on hormone and immune systems are considered hazardous. The Baltic Sea has been exposed to an extensive use of chemicals from the very beginning of the industrialization of the region in the late 19th century. Synthetic substances such as persistent organic pollutants (POPs) and pharmaceuticals and non-synthetic substances like heavy metals originate from: 1) point-sources of pollution situated on the coast or inland in the catchment areas
including industries and municipal wastewater plants; 2) land-based diffuse sources such as runoff from agricultural land, forests, and other land
uses as well as urban runoff, and leaching from dump sites and landfills;
10
3) activities taking place at sea such as shipping, pipelines, dredging or operation of oil platforms; and
4) atmospheric deposition of contaminants from all types of combustion sources as well as volatile chemicals (e.g., pesticides).
In the Baltic Sea region, industrial and municipal point sources have been most apparent sources of pollutants but several contaminant groups originate mainly from minor industrial sources, agriculture (e.g. pesticides, pharmaceuticals and fertilizers), households (e.g. various consumer products), sludge, dump sites and waste deposition in landfills (HELCOM 2010c). Atmospheric emissions come from land traffic, shipping, energy production, incineration of wastes and even small-scale household combustion. For some heavy metals (e.g. mercury, lead and cadmium), atmospheric deposition is a major component of their annual inputs to the Baltic Sea and for substances such as dioxins atmospheric deposition may dominate over other sources. These sources can be located outside the Baltic Sea region and it has been estimated that 60% of cadmium, 84% of lead and 79% of mercury deposited into the Baltic Sea originate from distant sources outside the Baltic Sea catchment area (mainly the UK, France, Belgium and Czech Republic) (Bartnicki et al. 2008). For dioxins this figure is 60 %. Disturbance of wildlife by hazardous substances has been documented for a number of trophic groups and the decline of grey and ringed seal populations has been attributed to the increase of organochloride substances. Although the populations of grey and ringed seals had already drastically declined due to hunting by the 1950s, the increasing pollution in the late 1960s by polychlorinated biphenyls (PCBs) from pulp and paper mills and other industrial sources caused a second decline which turned the populations to all-time lows (estimated at 5,000 ringed seals and 3,000 grey seals) by the 1970s (Harding et al. 1999). Organochlorides have been associated with reproductive failures in the seals. PCBs have been banned and since the mid-1980s the reproduction of seals has normalised, but as a very persistent contaminant they are still found in high concentrations in marine sediments and biota. Similar impacts have been shown on white-tailed eagle, mainly related to DDT, and other fish feeding birds. The adverse effects of DDT on the eggshell production of predatory birds and the consequent decline especially of white-tailed sea eagle populations in the northern hemisphere in the 1950s to 1970s were well demonstrated. White-tailed eagle populations recovered after the banning of DDT and PCBs and they, as well as smaller organisms, show today less contamination by PCBs. However, new contaminants like the estrogenic substances from municipal waste waters are causing new problems, such as feminization of male fish (Ferreira et al. 2009). In 1999–2007, the entire Baltic Sea was an area with a high contamination level. An indicator-based assessment of 144 areas in the Baltic Sea revealed that 137 areas were contaminated by hazardous substances (HELCOM 2010c). All open-sea areas of the Baltic Sea, except the north-western Kattegat, were classified as being “disturbed by hazardous substances”. The waters near larger coastal cities (Rostock, St. Petersburg, Helsinki, Gdańsk, Riga, Copenhagen and Stockholm) were generally classified as having a “moderate” or “poor” chemical status. Substances such as PCBs, lead, 1,1-dichloro-2,2-bis(p-
11
chlorophenyl)ethylene (DDE), cadmium, mercury, tributyltin (TBT), and dioxins as well as brominated substances, for example, brominated diphenylether (BDE) appeared as contaminants with the highest concentrations in relation to the threshold levels for disturbance. The above integrated status assessment did not address all hazardous substances that have been measured in Baltic fish or sediments. Such substances include, for example, all perfluorinated substances, alkyl phenyls, bisphenol A, and pharmaceutical substances. The concentrations of these substances in the marine environment are high and increasing. However, the understanding of their environmental fate is poor, and there is limited information about their spatial distribution, main sources and transport mechanisms.
15.3.4 Changes in biodiversity Diversity of species in the Baltic Sea is lower than diversity in most true marine areas. At present, about 6,000 species are known from the Baltic Sea and the Kattegat, including cyanobacteria, phytoplankton, zooplankton, phytobenthos, zoobenthos, fish, marine mammals and birds, as well as vertebrate parasites. Most species are either fresh or marine water species, and only one species, Fucus radicans, is a species endemic to the Baltic Sea (Ojaveer et al. 2010). Many freshwater and marine species live at their physiological limits in the brackish Baltic Sea. In addition, a low genetic diversity has also been uncovered in the Baltic. The low species and genetic diversity indicates that the biodiversity in the Baltic Sea may be particularly sensitive to disturbances. The biodiversity of the Baltic Sea has undergone major changes during the past decades. Especially, variations in climate, whether natural or anthropogenic, have introduced physical and chemical changes and consequently also biological shifts (HELCOM 2009b). While the lack of comprehensive data and the natural variability in biodiversity makes it challenging to identify the precise role of human pressures, they have no doubt contributed to the observed changes. HELCOM has identified species with a threatened or declining population status (HELCOM 2007b). The only species known to be extinct is the sturgeon (Acipenser sturio) but all marine mammals, except the grey seal north of 59° N latitude, are still under a threat or decline. The population of harbour porpoises (Phocoena phocoena), is in a perilous state with only a few hundred animals remaining in the Baltic Proper. The harbour porpoise was widely distributed and common until the early 20th century. Previously, the main cause of the decline was hunting but today the most important anthropogenic threats are incidental fisheries by-catch and prey depletion (HELCOM 2009b). In addition, factors such as hazardous substances and noise pollution are also likely to have negative impacts on the species. The grey seals (Halichoerus grypus) which were nearly hunted to extirpation in the beginning of the 20th century, are now recovering in the northern Baltic Sea, while south of 59° N the recovery is still very slow. The status of the ringed seal (Phoca hispida) is also still unfavourable. While the impacts of hunting and hazardous substances on seals have been reduced, fisheries by-catch and prey depletion are among the most prominent and continuing threats to these populations (HELCOM 2009b).
12
One of the most well-established changes in the Baltic ecosystem is the shift from dominance of demersal fish communities to pelagic clupeid fishes. The eastern Baltic cod (Gadus morhua) stock reached high abundances in the late 1970s and early 1980s (Eero et al. 2007) while in the 1980s, a climate-induced decrease in the cod reproductive volume, i.e., the amount of water with good conditions for hatching of cod eggs, caused high cod egg mortality (Köster et al. 2003). This, together with very high fishing pressure, resulted in historically low values of the cod stock since the early 1990s, however with some signs of improvement in the 2000s (ICES 2008). The clupeids, sprat (Sprattus sprattus) but also herring (Clupea harengus membras), on the other hand, have increased since the 1980s. Factors underlying the increase of especially sprat are the decreased predation by cod and possibly also eutrophication through an increase of food resources of the pelagic fish. In many coastal areas, fish which benefit from or tolerate eutrophication, such as percids and cyprinids have increased but in many other areas, fish stocks have overall declined owing to high fishing pressure. Several stocks of migratory fish species are also in a poor condition because of damming or blocking of migratory pathways or degradation of riverine habitat quality. Among the bird species, long-term population declines are evident for dunlin (Calidris
alpina schinzii), and more recently for common eider (Somateria mollissima) and long-tailed duck (HELCOM 2009b). The causes are not well understood, but climate change in the case of the dunlin, and shipping-induced oil spills, fisheries by-catch and habitat deterioration in the case of the ducks may have contributed to the decline. A recent threat assessment of Baltic breeding birds (HELCOM 2012b) records the gull-billed tern (Gelochelidon nilotica) as regionally extinct. Moreover, the kentish plover (Charadrius alexandrines) is critically endangered and dunlin, terek sandpiper (Xenus cinereus) and black-legged kittiwake (Rissa
tridactyla) endangered. The development of bird populations also includes positive signs such as the recovery of the white-tailed eagle (Haliaeetus albicilla) and great cormorant (Phalacrocorax carbo sinensis). Macrobenthic soft-sediment communities are currently severely degraded and abundances are below a 40-year average in the entire Baltic Sea (HELCOM 2009b). The increased prevalence of oxygen-depleted deep water is perhaps the single most central factor influencing the structural and functional biodiversity of benthic communities in the open-sea areas of the Baltic Sea. The zooplankton community has as already mentioned also experienced significant changes over recent decades particularly in the offshore copepod communities in the Baltic Proper and the southern Baltic Sea (HELCOM 2009b). Changes in salinity and temperature are likely important factors underlying observed changes. In addition, eutrophication has contributed to the decreasing volume of oxygenated water below the halocline in offshore areas, thereby reducing the volume of water suitable for the reproduction of zooplankton species that require higher salinities. Nutrient enrichment has resulted in increased phytoplankton productivity, including more prevalent algal blooms and which themselves can be considered as a manifestation of reduced biodiversity within the phytoplankton community (HELCOM 2009b). A number of
13
changes in the community composition have also occurred during the past thirty years, e.g., a shift in dominance from diatoms to dinoflagellates during spring bloom periods. In the macrophyte communities, important habitat forming species such as bladder wrack, eelgrass, and stoneworts have decreased in abundance and distribution in many coastal areas, including a decrease of the depth penetration of bladder wrack (Fucus vesiculosus) (Kautsky et al. 1986). This has been primarily attributed to reduced water transparency. The decrease in all habitat builders is most pronounced in highly polluted and eutrophied areas as well as areas subject to physical disturbance to the bottom (HELCOM 2009). Decline in habitat forming species has implications beyond the local scale since they are important living, feeding, reproduction and nursing environments for associated flora and fauna. More recently submerged aquatic vegetation is showing signs of recovery e.g., in the north-western and north-eastern Baltic Proper. Alien species are both a part of and a threat to the Baltic Sea biodiversity. Since the early 1800s, about 120 alien species have been introduced to the Baltic Sea including the Kattegat. Most of the introduced species are crustaceans (23 species) and the main introduction pathway has been shipping (HELCOM 2009b). The benthic bristle worm Marenzelleria spp. presents one of the best documented cases of invasive species. It took the species roughly ten years to spread to the entire Baltic Sea and to become common in many soft-bottom habitats. Often the ecological impacts of alien species on the Baltic Sea ecosystem have been difficult to observe or they are poorly understood. In the most heavily invaded coastal lagoons of the southern Baltic, several food chains and even major parts of sea-bottom communities may be based on introduced species (Leppäkoski et al. 2002). The increasing spread of alien species increases the risk that native species or habitats of high conservation value will eventually be impacted. A pilot study using indicators of biodiversity and an integrated assessment approach was carried out in 73 open-sea and coastal areas for the time period 2003–2007 (HELCOM 2010). According to these results, 82% of the coastal areas assessed had an unfavourable conservation status and only 18% were in a ‘good’ or ‘high’ status. In terms of ecosystem health, the deterioration of the status of biodiversity is critical because it diminishes the resilience or buffering capacity against large-scale shifts in the Baltic Sea ecosystem and increases the risk for escalating degradation of the environment (HELCOM 2009b).
15.3.5 Holistic assessment of ecosystem health An initial holistic assessment of ecosystem health of the Baltic Sea in 2003-2007 (Figure 15.4) showed that the whole Baltic Sea marine area, except one site in the Gulf of Bothnia, was classified as having a moderate, poor or bad ecosystem health (HELCOM 2010a). The only areas close to having “good ecosystem health status” include the Bothnian Bay, the Bothnian Sea and parts of the northern Kattegat. The Baltic Sea ecosystem is degraded to such an extent that its capacity to deliver ecosystem goods and services to the people living in the nine coastal states has been reduced. These results were based on chemical, biological and supporting indicator data from altogether 84 assessment areas (HELCOM 2010a). The state of open waters was in some areas worse than that of the coastal waters (Figure 15.3). In the Baltic Sea Pressure Index open sea areas especially in the southern and eastern
14
areas were under a high sum of pressures from a number of sources (Figure 15.2B). The status of the ecosystem in the open sea was heavily affected by open sea fisheries and inputs of nutrients and hazardous substances. In addition, the poor status of sediments of the open sea areas resulting from the accumulation of hazardous substances or hypoxia was in many cases the underlying factor behind the poor ecosystem health status.
Figure 15.3 Ecosystem health of the Baltic Sea in 2003-2007 assessed using biological, chemical and
supporting indicators. Blue colour indicates high status, green good, yellow moderate, orange poor and red bad
status in coastal (small dot) or open sea (large dot) assessment sites. From HELCOM (2010a).
15.4 Is the Baltic Sea a healthy ecosystem?
15.4.1 Concept of ecosystem health Ideally, a healthy ecosystem is an ecosystem with full functionality and potential, and with the absence of distress symptoms caused by anthropogenic stressors (Rapport 2007). An
15
analogy can be drawn between humans and ecosystems: just like an increase in human body temperature or signs of inflammation are symptoms of human illness, declines in populations of top predators, algal blooms or bottom hypoxia can be considered symptoms of deteriorating health of a marine ecosystem. Ecosystem vitality on the other hand is reflected in nutrient and energy flows, biodiversity and the capacity of the ecosystems to rebound from perturbations (Rapport 2007). Historical data probably provides the most appropriate means to evaluate whether symptoms of deteriorating ecosystem health are present. It enables the analysis of trends in ecosystem state and deterioration and the comparison of current ecosystem state to a pristine or less impacted state in the past. Alternatively, pristine sites or areas within a marine region can be used as reference sites for a healthy ecosystem. However, such pristine sites are no longer found in the Baltic Sea. This kind of an ecosystem health approach to human impacts was highly relevant at the time when ecosystems were not yet complexly impacted and impacts from human activities only started to become increasingly apparent. More recently, the attention has turned to estimating how much anthropogenic impact to ecosystems is acceptable and how the ecosystems can be restored if the boundaries of acceptable levels have been surpassed.
15.4.2 Sustainable use of the marine ecosystem Today, humans are seen as an integral part of the ecosystems and negative impacts from human presence and resource use are considered unavoidable and to some extent acceptable. Desirable states of marine ecosystems are increasingly defined from the point of view of human needs, and environmental protection is implemented through the concepts of sustainable development and sustainable use of ecosystems. The basis for the concept of sustainable development was laid already by the 1972 United Nations Conference on the Human Environment. It brought to the wider global political attention the need to protect the environment while pursuing development. The concept of sustainable development was reinvigorated by the 1992 United Nations Rio Conference on Environment and Development which emphasized the right to development of present and future human generations, as well as the sovereign right of states to exploit their own resources, but in such a way that environmental protection will constitute an integral part of the development process. Sustainable development as it is interpreted today aims at balancing between the current economic, social and environmental interests without compromising the well-being of future generations. Today the ecosystem approach and ecosystem-based management of human activities form the basis of environmental policies related to the marine environments. The twelve “Malawi principles of the ecosystem approach to biodiversity management” presented to the Fourth Meeting of the Parties of the UN Convention on Biological Diversity in 1998 are often referred to as the starting point for the ecosystem approach. These principles underline for example that conservation of ecosystem structure and functioning is important, that appropriate balance between conservation and use of biodiversity should be pursued, and most importantly ecosystems must be managed within the limits of their functioning. Both the HELCOM Baltic Sea Action Plan as well as the EU Marine Strategy Framework Directive
16
(MSFD), both of which are highly relevant for the Baltic Sea, are based on the ecosystem approach to the management of human activities. Good understanding of the ecosystem structure and functions and of the pathways and mechanisms through which human activities impact the ecosystem are fundamental. Ideally, the pressure – impact relationships would be understood and presented by response curves (cf. Figure 15.1) which would guide managers to predict a sustainable amount of pressure. However, currently we are far from such precise understanding and marine assessments are forced to simplify the reality. Defining what is a “healthy ecosystem” and sustainable level of use for a particular marine ecosystem requires not only good scientific understanding of the ecosystem’s structure and functioning and responses to anthropogenic pressures. More importantly it requires value judgments. The advice from science is ultimately operationalized through political decisions related to how much degradation of the environmental quality of the ecosystem the society is ready to accept (e.g. Mee et al. 2008). Recently, the concepts of ecosystem thresholds, resistance or resilience, as well as carrying capacity have become centrepieces in defining acceptable levels of pressures and impacts resulting from human activities (e.g. Elliott et al 2007). Due to the need to keep on the right side of the pressure thresholds for regime shifts and the incomplete knowledge on the ways on which the pressures act on marine systems, there is a need to increase knowledge on how these concepts materialise in reality (Mee et al. 2008). In the Baltic Sea, it took nearly four decennia from the recognition of the DDT problem to the recovery of the white-tailed eagle populations through political decisions to ban the substance (Elmgren 2001). Similarly, it took some decennia from the recognition of the mercury and PCB contamination problem to the recovery but the politicians did not wait until scientific consensus to emerge before they took decisions to ban those substances (Elmgren 2001). Precautionary and adaptive management approaches have been called for by the scientists when the not-yet-so-well-known environmental problems are to be managed (Mee et al. 2008, Elmgren 2001). This means that the society should not wait until scientific consensus emerges and act only then but instead take action already in the presence of uncertainty.
15.4.3 Is the present level of human use of the Baltic Sea sustainable? The Baltic Sea has not been managed in a sustainable manner and the use of the ecosystem goods and services has surpassed the boundaries of sustainable use. In the assessment of the sum of human pressures on the Baltic Sea, the whole Baltic Sea was under some level of pressure from human activities and the pressures were greatest in the eastern and southern areas (HELCOM 2010a, 2011). The thematic assessments showed that the levels of eutrophication, pollution by hazardous substances and degradation of biodiversity were unacceptable in most of the Baltic Sea area (HELCOM 2009a, 2009b and 2010c). Similarly, the holistic assessment of ecosystem health demonstrated that in only one pocket in the coastal area of the northern Bothnian Sea a good ecosystem health status could be found (HELCOM 2010a).
17
In addition to these assessments, the prevalence of regime shifts caused partly by overfishing and eutrophication provides first hand evidence that the levels of sustainable use of the ecosystem have been surpassed for the Baltic Sea ecosystem (Österblom et al. 2007). It is likely that some ecosystem impacts can be reverted, whilst others cannot. Positive examples include those from coastal areas where the introduction or upgrading of waste water treatment has led to the improvement of the water quality and recovery of the ecosystem status (e.g. Finni et al. 2001). Recoveries of the northern Baltic grey seal populations, as well as the resurge of populations of great cormorants and white-tailed eagles, after the relaxation of hunting and banning of the use of organochlorides DDT and PCBs are also among the recovery success stories. Reverting to an acceptable eutrophication status and restoring the Baltic Sea biodiversity will be much more challenging if not unachievable since there are now feedback mechanisms in play such as internal loading of phosphorus from sediments and prevalence of nitrogen-fixing cyanobacteria that maintain the eutrophicated state and have led to the potentially inhibited recovery (Vahtera et al. 2007). As concerns the biodiversity of the Baltic Sea, the communities and food webs of the Baltic Sea have changed as a result of climatic changes, fishing and eutrophication and the ecosystem has reorganised through regime shifts. Due to the profound changes both in the underlying physical and chemical factors as well as the populations of species, it is highly unlikely that it will be possible to return through active management the Baltic Sea ecosystem to a state it had before heavy human influence e.g. in the beginning of the 20th century. Recent reports have demonstrated that nutrient loads to the Baltic Sea have declined (HELCOM 2011) and that there are decreasing concentrations of nutrients in the sea, but eutrophication indicators such as chlorophyll a levels still do not shown signs of decrease (HELCOM 2012a). Overall, the return paths to healthier states of the ecosystem are still largely unknown for the Baltic Sea (cf. Figures 15.1A to D). When considering eutrophication, it is clear that due to the long memory of the system, e.g. long residence time of water, there are time lags involved in the recovery and the recovery will not result in a similar ecosystem that was present before large scale eutrophication. This will result in a recovery trajectory with a shifted baseline (Fig. 15.1B and D). Even if there is no return to a previous state, there is the obligation in the current policies to manage the ecosystem prevailing physical and climatic conditions and the requirement to ensure that the thresholds of current regime will not be exceeded.
15.5 Marine ecosystem health objectives in
international policies
15.5.1 The aim: “Good environmental status” At the global level, the 1982 UN Convention of the Law of the Sea (UNCLOS) determines that states have the obligation to protect and preserve the marine environment but UNCLOS does not define more specific goals for the desired state.
18
At the European level the EU Marine Strategy Framework Directive (MSFD) adopted in 2008 is the most explicit piece of EU environmental legislation addressing the status of marine environments (Anon 2008a). The MSFD aims at measures that will yield “good environmental status” of the marine waters. It gives a description of what is a “good environmental status”, a concept that also encompasses ecosystem status. The MSFD stipulates that "good environmental status" means the environmental status of marine waters where these “provide ecologically diverse and dynamic oceans and seas which are
clean, healthy and productive within their intrinsic conditions, and the use of the marine
environment is at a level that is sustainable, thus safeguarding the potential for uses and
activities by current and future generations”. The MSFD further reads that good environmental status is achieved when: “The structure, functions and processes of the
constituent marine ecosystems, together with the associated physiographic, geographic,
geological and climatic factors, allow those ecosystems to function fully and to maintain their
resilience to human-induced environmental change. Marine species and habitats are
protected, human-induced decline of biodiversity is prevented and diverse biological
components function in balance. Hydro-morphological, physical and chemical properties of
the ecosystems, including those properties which result from human activities in the area
concerned, support the ecosystems as described above. Anthropogenic inputs of substances
and energy, including noise, into the marine environment do not cause pollution effects.” This definition has been complemented by eleven qualitative descriptors of good environmental status (Annex 1 of the Directive), and 26 criteria and 56 sub-criteria that should be considered when assessing the state of the sea (Anon 2010). The Directive obliges the member states to develop and implement marine strategies which aim at reaching the good environmental status.
15.5.2 The aim: “Good ecological status” The EU Water Framework Directive (WFD) addresses primarily fresh water environments but also river mouths and marine coastal waters extending one (for ecological status) or twelve (for chemical status) nautical miles seawards from the baselines, and sets an objective to maintain or move towards “good ecological status” (Anon 2000). The directive stipulates what should be assessed when evaluating the status, so called “biological quality elements”, and provides normative definitions by words for what is a good ecological status. For coastal waters, factors such as composition and abundance of phytoplankton, other aquatic flora and benthic invertebrate fauna need to be assessed together with supporting hydromorphological, chemical and physico-chemical elements. In general, good status for these quality elements is described as a status with only slight changes caused by human pressures from the reference conditions that are defined for different water-types and for each quality element.
15.5.3 The aim: “Favourable conservation status of species and habitats” The UN Convention on Biological Diversity (CBD) is the most important agreement addressing biodiversity protection at the global level. Global targets have been set for improving the state of biodiversity, including marine biodiversity, under, e.g. the so called Aichi biodiversity targets.
19
At the European level, Habitats and Birds Directives have set as an objective to reach favourable conservation status of species and habitats listed in these directives (Anon 1992, 2009). Some of the species and habitats on these lists are also found in the Baltic Sea, such as species of marine mammals, seabirds and underwater marine habitats like reefs, estuaries and lagoons. According to the Habitats Directive, the conservation status of a natural habitat is ‘favourable’ when its natural range and areas it covers within that range are stable or increasing, the specific structure and functions which are necessary for its long-term maintenance exist and are likely to continue to exist for the foreseeable future, and the conservation status of its typical species is favourable. The conservation status of species is in turn ‘favourable’ when population dynamics data on the species concerned indicate that it is maintaining itself on a long-term basis as a viable component of its natural habitats, and the natural range of the species is neither being reduced nor is likely to be reduced for the foreseeable future, and there is, and will probably continue to be, a sufficiently large habitat to maintain its populations on a long-term basis. Furthermore, the Habitat Directive requires the designation of a network of protected areas, labelled Natura 2000 that hosts the habitats and species listed in the annexes of the Directive. The Birds Directive is the oldest EC nature conservation directive and aims at protecting naturally occurring bird species in Europe. It applies to birds, and their nests and eggs, and particularly focuses on the protection of habitats for endangered as well as migratory species. This directive also requires the designation of protected areas that nowadays are part of the Natura 2000 network.
15.5.4 Regional cooperation for the better health of the Baltic Sea The global or European policies and their environmental objectives do not specifically address the Baltic Sea ecosystem. However, their implementation requires interpretation and application at the Baltic Sea level so that we know what the jurisdictional language means when the species, habitats and environmental conditions of the Baltic Sea are in focus. Governments of the Baltic Sea coastal countries have worked together to protect the marine environment of the Baltic Sea since 1974 when the Convention for the Protection of the Marine Environment of the Baltic Sea was first signed. Today Helsinki Commission (HELCOM), which is the governing body of the Convention, gathers together the coastal countries and the EU, represented by the European Commission, to discuss and agree on what are the goals and measures needed for a healthy Baltic Sea. HELCOM also provides a platform for regional coordination: to consider how the EU Directives’ jurisdictional language is applied when it comes to the ecosystem of the Baltic Sea and how the implementation of the legislation can be coordinated at the Baltic Sea level. The Baltic Sea Action Plan (BSAP) was adopted in 2007 by the ministers and high-level representatives of the HELCOM Contracting Parties (HELCOM 2007, Backer et al. 2010).
20
With the BSAP the coastal countries of the Baltic Sea committed themselves to reaching a healthy Baltic Sea ecosystem by year 2021 through applying the actions outlined in the Action Plan. The ultimate vision of HELCOM is to reach a healthy Baltic Sea environment, with diverse biological components functioning in balance, resulting in a good environmental/ecological status and supporting a wide range of sustainable human economic and social activities. The plan focuses on reducing eutrophication, contamination by hazardous substances, enhancing biodiversity and nature conservation, as well as securing environmentally friendly maritime activities. For reducing eutrophication, the BSAP contains a full-fledged critical load approach with maximum allowable loads of nitrogen and phosphorus defined for each Baltic Sea sub-basin and country based on ecological targets for eutrophication ecological modelling. The maximum allowable inputs are accompanied by provisional nutrient load reduction requirements for each coastal country. This scheme is the first internationally agreed targets-based nutrient load reduction scheme for an entire regional sea (Backer et al. 2010).
References Alheit J, Möllmann C, Dutza J, Kornolovs, G, Loew P, Mohrholz V, Wasmund N (2005) Synchronous ecological
regime shifts in the central Baltic and the North Sea in the late 1980s. ICES Journal of Marine Science 62: 1205-1215
Andersen JH, Axe P, Backer H, Carstensen J, Claussen U, Fleming-Lehtinen V, Järvinen M, Kaartokallio H, Knuuttila S, Korpinen S, Kubiliute A, Laamanen M, Lysiak-Pastuszak E, Martin G, Møhlenberg F, Murray C, Møhlenberg F, Nausch G, Norkko A, Villnäs A (2010) Getting the measure of eutrophication in the Baltic Sea: towards improved assessment principles and methods. – Biogeochemistry, DOI 10.1007/s10533-010-9508-4
Anon (1992) Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal L 206
Anon (2000) Directive 200/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Official Journal of the European Communities L 327/1
Anon (2008a) Directive 2008/56/EC of the European Parliament and of the Council of 17 June 2008 establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive). Official Journal of the European Communities L 164/19
Anon (2008b) - Directive 2008/105/EC of the European Parliament and of the Council of 16 December 2008 on environmental quality standards in the field of water policy, amending and subsequently repealing Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and amending Directive 2000/60/EC of the European Parliament and of the Council. Official Journal of the European Communities L 348
Anon (2009) Directive 2009/147/EC of the European Parliament and of the Council of 30 November 2009 on the conservation of wild birds (codified version). Official Journal of the European Union L20/7
Anon (2010) Commission decision of 1 September 2010 on criteria and methodological standards on good environmental status of marine waters. Official Journal of the European Union L232
Backer H, Leppänen JM, Brusendorff AC, Forsius K, Stankiewicz M, Mehtonen J, Pyhälä M, Laamanen M, Paulomäki H, Vlasov N, Haaranen T (2010) HELCOM Baltic Sea Action Plan – A regional programme of measures for the marine environment based on the Ecosystem Approach. Mar Pol Bull 60: 642-649
Bartnicki J, Gusev A, Aas W, Fagerli H, Valiyaveetil S (2008): Atmospheric supply of nitrogen, lead, cadmium, mercury and dioxins/furans to the Baltic Sea in 2006. EMEP report to HELCOM. Online 30 December 2011, http://www.helcom.fi/environment2/hazsubs/EMEP/en_GB/emep2006/
Bartnicki J (2011) Atmospheric nitrogen depositions to the Baltic Sea during 1995-2009. HELCOM Indicator Fact Sheets 2011. Online 29 December 2011, http://www.helcom.fi/environment2/ifs/en_GB/cover/
Conley DJ, Humborg C, Rahm L, Savchuk OP, Wulff F (2002) Hypoxia in the Baltic Sea and basin-scale changes in phosphorus biogeochemistry. Environ. Sci. and Technol. 15:5315-5320
Duarte CM, Conley DJ, Carstensen J, Sánchez-Camacho M (2009) Return to Neverland: Shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32:29-36
21
Dippner JW, Vuorinen I, Daynus D, Flinkman J, Halkka A et al (2008). Climate-related marine ecosystem change. In: The BACC Author Team, Assessment of Climate Change for the Baltic Sea Basin. Springer-Verlag, Berlin, Heidelberg. 473 pp
Eero M, Koster FW, Plikshs M, Thurow F (2007) Eastern Baltic cod (Gadus morhua callarias) stock dynamics: extending the analytical assessment back to the mid-1940s. ICES Journal of Marine Science 64: 1257-1271
Elliott M, Burdon D, Hemingway KL, Apitz SE (2007) Estuarine, coastal and marine ecosystem restoration: Confusing management and science – A revision of concepts. Estuar Coast Shelf Sci 74: 349-366
Elmgren R (2001) Understanding Human Impact on the Baltic Ecosystem: Changing Views in Recent Decades. Ambio 30:222-231
Ferreira F, Santos MM, Filipe L, Castro C, Reis-Henriques MA, Lima D, Vieira MN, Monteiro NM (2009) Vitellogenin gene expression in the intertidal blenny Lipophrys pholis: a new sentinel species for estrogenic chemical pollution monitoring in the European Atlantic coast? Comp. Biochem. Physiol. C. Toxicol. Pharmacol. 149:58-64
Finni, T, Laurila S, Laakkonen S (2001) The history of eutrophication in the sea area of Helsinki in the 20th Century long-term analysis of plankton assemblages. Ambio 30:264-271
Fleming-Lehtinen V, Kauppila P, Kaartokallio H (2010): How far are we from clear water? HELCOM Indicator Fact Sheets 2011. Online 30 December 2011, http://www.helcom.fi/BSAP_assessment/eutro/Secchi/en_GB/status/
Hansson S. Hjerne O, Harvey C, Kitchell JF, Cox SP, Essington TE (2007) Managing Baltic Sea fisheries under contrasting production and predation regimes – ecosystem model analyses. Ambio 36:265–271
Harding KC, Härkönen TJ (1999) Development in the Baltic grey seal (Halichoerus grypus) and ringed seal (Phoca hispida) populations during the 20th century. Ambio 28:619-627
Heino R, Tuomenvirta H, Vuglinsky VS, Gustafsson BG, Alexandersson H et al (2008). Past and current climate change. In: The BACC Author Team. Assessment of Climate Change for the Baltic Sea Basin. Springer-Verlag, Berlin, Heidelberg. 473 pp
HELCOM (2007a) Baltic Sea Action Plan. Helsinki Commission. 103 pp. http://www.helcom.fi/stc/files/BSAP/BSAP_Final.pdf
HELCOM (2007b) HELCOM List of threatened and/or declining species and biotopes/habitats in the Baltic Sea area. Balt. Sea Environ. Proc. No. 113. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep113.pdf
HELCOM (2009a) Eutrophication in the Baltic Sea – An integrated thematic assessment of the effects of nutrient enrichment in the Baltic Sea region. Balt. Sea Environ. Proc. No. 115B. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep115B.pdf
HELCOM (2009b) Biodiversity in the Baltic Sea - An integrated thematic assessment on biodiversity and nature conservation in the Baltic Sea. Balt. Sea Environ. Proc. No. 116B. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep116B.pdf
HELCOM (2010a) HELCOM Initial Holistic Assessment of the Ecosystem Health of the Baltic Sea 2003-2007. Balt. Sea Environ. Proc. No. 122. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep122.pdf
HELCOM (2010b) Towards a tool for quantifying the anthropogenic pressures and potential impacts in the Baltic Sea marine environment: A background document on the method, data preparation and testing of the Baltic Sea Pressure and Impact Indices. Balt. Sea Environ. Proc. No. 125. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep125.pdf
HELCOM (2010c) Hazardous substances in the Baltic Sea - An integrated thematic assessment of hazardous substances in the Baltic Sea. Balt. Sea Environ. Proc. 120B. Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/bsep120B.pdf
HELCOM (2011) The Fifth Baltic Sea Pollution Load Compilation (PLC-5) Balt. Sea Environ. Proc. No. 128 Helsinki Commission. http://www.helcom.fi/stc/files/Publications/Proceedings/BSEP128.pdf
HELCOM (in print) Approaches and methods for eutrophication target setting in the Baltic Sea region. Balt. Sea Environ. Proc. No. ## Helsinki Commission.
HELCOM (2012b) HELCOM Red List of Species and Habitats/Biotopes: Red List of Baltic Breeding Birds, version 2011. Helsinki Commission.
ICES (2008): Report of the Baltic Fisheries Assessment Working Group (WGBFAS), 8-17 April 2008. ICES CM 2008/ACOM: 06, 692 pp
22
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, Bourque BJ, Bradbury RH, Cooke R, Erlandson J et al. (2001) Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629-638
Kautsky N, Kautsky H, Kautsky U, Waern M (1986) Decreased depth penetration of Fucus vesiculosus (L.) since the 1940’s indicates eutrophication of the Baltic Sea. Mar. Ecol. Prog. Ser. 28:1-8
Kemp WM, Testa JM, Conley DJ, Gilbert D, Hagy JD (2009) Temporal responses of coastal hypoxia to loading and physical control. Biogeosciences 6:2985-3008
Korpinen S, Meski L, Andersen JH, Laamanen M (2011): Human pressures and their potential impact on the Baltic Sea ecosystem. Ecol Ind 15: 105-114
Köster FW, Hinrichsen H, Schnack D et al. (2003) Recruitment of Baltic cod and sprat stocks: identification of critical life stages and incorporation of environmental variability into stock-recruitment relationships. Scientia marine 67 (Suppl. 1): 129-154
Leppäkoski E, Gollasch S, Gruszka P, Ojaveer H, Olenin S, Panov V. (2002) The Baltic - a sea of invaders. Can J Fish Aquat Sci 59: 1175-1188
Mee L, Jefferson RL, Laffoley Dd’A, Elliott M (2008) How good is good? Human values and Europe’s proposed Marine Strategy Directive. Mar Pol Bull 56: 187-204
Nixon SW (1995) Coastal marine eutrophication: a definition, social causes, and future concerns. Ophelia 41:199-219
Ojaveer H, Jaanus A, MacKenzie BR, Martin G, Olenin S,5, Radziejewska T, Telesh I. Zettler M, Zaiko A (2010) Status of Biodiversity in the Baltic Sea. PLoS ONE 5(9): e12467
Österblom H, Hansson S, Larsson U, Hjerne O, Wulff F, Elmgren R, Folke C (2007) Human-induced trophic cascades and ecological regime shifts in the Baltic Sea. Ecosystems 10:877-889
Rapport D (2007) Sustainability science: an ecohealth perspective. Sustain Sci. doi:10.1007/s11625-006-0016-3
Savchuk OP, Wulff F, Hille S, Humborg C, Pollehne F (2008) The Baltic Sea a century ago - a reconstruction from model simulations, verified by observations. J of Mar Systems 74:485-494
Scheffer M, Carpenter S, Foley JA, Folke C, Walker B (2001) Catastrophic shifts in ecosystems. Nature 413:591-596
Schmidt-Thomé P (ed.) (2006). Sea Level Change Affecting the Spatial Development of the Baltic Sea Region. Geological Survey of Finland, Special Paper 41, 154 pp
Struck U, Emeis K, Voss M, Christiansen C, Kunzendorf H (2000): Records of southern and central Baltic Sea eutrophication in δ13C and δ15N of sedimentary organic matter. Mar. Geol. 164:157-171
Thurow F (1997) Estimation of the total fish biomass in the Baltic Sea during the 20th century. ICES J. of Mar. Sci. 54:444-461
Vahtera E, Conley DJ, Gustafsson BG, Kuosa H, Pitkänen H, Savchuk OP, Tamminen T, Viitasalo M, Voss M, Wasmund N, Wulff F (2007) Internal ecosystem feedbacks enhance nitrogen-fixing cyanobacteria blooms and complicate management in the Baltic Sea. Ambio 36:186-194
Winsor P., Rodhe J, Omstedt A (2001) Baltic Sea ocean climate: an analysis of 100 yr of hydrographic data with focus on the freshwater budget. Clim. Res. 18:5-15
Zillén L & Conley DJ (2010) Hypoxia and cyanobacteria blooms - are they really natural features of the late Holocene history of the Baltic Sea? Biogeosciences 7:2567-2580
