Making the ecosystem approach operational—Can regime shifts in ecological- and governance systems facilitate the transition?

Marine Policy 34 (2010) 1290–1299

Contents lists available at ScienceDirect

Marine Policy

0308-59

doi:10.1

n Corr

Stockho

fax: +4

E-m

journal homepage: www.elsevier.com/locate/marpol

Making the ecosystem approach operational—Can regime shiftsin ecological- and governance systems facilitate the transition?

H. Osterblom a,b,n, A. Gardmark c, L. Bergstrom c, B. Muller-Karulis d, C. Folke b,e, M. Lindegren f,M. Casini g, P. Olsson b, R. Diekmann h, T. Blenckner a, C. Humborg a, C. Mollmann h

a Baltic Nest Institute, Stockholm Resilience Centre, Stockholm University, SE-106 91 Stockholm, Swedenb Stockholm Resilience Centre, Stockholm University, SE-106 91 Stockholm, Swedenc Institute of Coastal Research, Swedish Board of Fisheries, Skolgatan 6, SE-742 42 Oregrund, Swedend Latvian Institute of Aquatic Ecology, 8 Daugavgrivas, LV-1048 Riga, Latviae Beijer Institute of Ecological Economics, The Royal Swedish Academy of Sciences P.O. Box 50005, SE-104 05 Stockholm, Swedenf National Institute of Aquatic Resources, Technical University of Denmark, Charlottenlund Slot, Charlottenlund DK-2920, Denmarkg Institute of Marine Research, Swedish Board of Fisheries, P.O. Box 4, SE 453 21 Lysekil, Swedenh Institute for Hydrobiology and Fisheries Science, University of Hamburg, Grosse Elbstrasse 133, D-22767 Hamburg, Germany

a r t i c l e i n f o

Article history:

Received 12 April 2010

Received in revised form

27 May 2010

Accepted 27 May 2010

Keywords:

Diffusion of innovation

Ecosystem approach

Marine governance

Regime shift

Resilience

7X/$ - see front matter & 2010 Elsevier Ltd. A

016/j.marpol.2010.05.007

esponding author at: Baltic Nest Institute, S

lm University, SE-106 91 Stockholm, Sweden

6 8 674 70 20.

ail address: henrik.osterblom@stockholmresil

a b s t r a c t

Effectively reducing cumulative impacts on marine ecosystems requires co-evolution between science,

policy and practice. Here, long-term social–ecological changes in the Baltic Sea are described,

illustrating how the process of making the ecosystem approach operational in a large marine ecosystem

can be stimulated. The existing multi-level governance institutions are specifically set up for dealing

with individual sectors, but do not adequately support an operational application of the ecosystem

approach. The review of ecosystem services in relation to regime shifts and resilience of the Baltic Sea

sub-basins, and their driving forces, points to a number of challenges. There is however a movement

towards a new governance regime. Bottom-up pilot initiatives can lead to a diffusion of innovation

within the existing governance framework. Top-down, enabling EU legislation, can help stimulating

innovations and re-organizing governance structures at drainage basin level to the Baltic Sea catchment

as a whole. Experimentation and innovation at local to the regional levels is critical for a transition to

ecosystem-based management. Establishing science-based learning platforms at sub-basin scales could

facilitate this process.

& 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Complex ecosystems continuously change and managing fortheir resilience (their capacity to absorb disturbance and reorga-nize while undergoing change, in order to retain their structureand function) [1] requires an adaptive governance strategy [2].Adaptive governance conveys the difficulty of control, the need toproceed in the face of uncertainty, and the importance of dealingwith diversity and conflict among stakeholders, who differ invalues, interests, perspectives and power [3]. Such governancerequires co-ordination that enables self-organization and adaptiveco-management of ecosystems [4]. For such governance to beeffective, an understanding of both ecosystem dynamics andsocial–ecological interactions is needed [5].

ll rights reserved.

tockholm Resilience Centre,

. Tel.: +46 8 674 76 64;

ience.su.se (H. Osterblom).

Governance systems designed to deal with complexity oftenrely on multi-level arrangements where authority has beenreallocated upward, downward and sideways away from centralstates [6–8]. It has been proposed that such diverse structures canaddress environmental problems at multiple scales and nurturediversity for dynamic responses, thereby complementing topdown, command and control management [9–11]. In policy terms,there are many similarities between adaptive co-management andthe ecosystem approach to management, as defined in theConvention of Biological Diversity [12] (www.cbd.int/ecosystem,Table 1). In principle, the ecosystem approach entails that scales ofmanagement should be matched to relevant ecological scales, inorder to manage for maintained structure, function and resilience.

The ecosystem approach is commonly featured in marinepolicy documents, but managers commonly struggle with itsinterpretation and practical implementation [13–15]. It is notuncommon to apply a narrow definition of the concept, focusingon the effects of fishing on non-target species, or other food web-related issues [16]. This ‘‘food web approach’’ often lacks theimportant human dimension [17] thereby only involving parts of

Table 1Malawi principles for the ecosystem approach [2].

(1) Management objectives are a matter of societal choice.

(2) Management should be decentralized to the lowest appropriate level.

(3) Ecosystem managers should consider the effects of their activities on

adjacent and other ecosystems.

(4) Recognizing potential gains from management there is a need to understand

the ecosystem in an economic context, considering e.g., mitigating market

distortions, aligning incentives to promote sustainable use and internalizing

costs and benefits.

(5) A key feature of the ecosystem approach includes conservation of ecosystem

structure and functioning.

(6) Ecosystems must be managed within the limits to their functioning.

(7) The ecosystem approach should be undertaken at the appropriate scale.

(8) Recognizing the varying temporal scales and lag effects which characterize

ecosystem processes, objectives for ecosystem management should be set for

the long term.

(9) Management must recognize that change is inevitable.

(10) The ecosystem approach should seek the appropriate balance between

conservation and use of biodiversity.

(11) The ecosystem approach should consider all forms of relevant information,

including scientific and indigenous and local knowledge, innovations and

practices.

(12) (12) The ecosystem approach should involve all relevant sectors of society

and scientific disciplines.

H. Osterblom et al. / Marine Policy 34 (2010) 1290–1299 1291

the identified elements of an ecosystem approach (Table 1).Furthermore, the appreciation of complex systems and potentialregime shifts is yet to be incorporated in marine ecosystem-basedmanagement [18]. The ecosystem approach has to involve a co-evolution between science and policy [19] and also emphasize theimportance of governance1 [20,21].

The ecosystem approach is knowledge intensive and requires athorough understanding of ecosystem structure and function, thedynamics of ecosystem services and their driving forces, spatialresilience of sub-systems within the ecosystem and responsediversity of species and functional groups. The literature onecosystem-based management, e.g., [22,23] is only recently[24–27] starting to include empirically based insights intostrategies that make transitions to such management possible.

Here, we attempt to describe the social–ecological process ofmaking the ecosystem approach operational. The Baltic Sea, alarge marine ecosystem with multiple governance structures, isused as an example. The questions investigated are:

(1)

1

that

oppo

forb

[20].

How can current understanding of ecosystem dynamicscontribute to operational ecosystem-based management(given the current governance framework)?

(2)

The aim of this synthesis is to contribute to the understandingof how adaptive ecosystem governance regimes can evolve in alarge marine ecosystem, and to identify governance processesthat facilitate implementation of an ecosystem approach.

2. Ecological knowledge for understanding and managingcomplex ecosystems

According to the ecosystem approach (Table 1), the objectivesfor management should be a matter of societal choices. However,

Defined as the formal and informal arrangements, institutions and mores

structure how resources or an environment are utilized, how problems and

rtunities are evaluated and analyzed, what behavior is deemed acceptable or

idden, and what rules and sanctions are applied to affect the pattern of use

an overall objective is the preservation of ecosystem structure andfunction (Table 1). Ecological knowledge required for ecosystem-based management includes an understanding of ecosystemstructure (patterns) and dynamics (processes), the functions thatthe ecosystem provides (goods and services), and ecosystemresponses to change, such as environmental variation andanthropogenic pressures (addressing the issue of resilience).Where differences in priorities between geographic scales andinterest groups are expected, scientific understanding needs to beclear about ecosystem interactions and potentially conflictingobjectives, as well as the appropriate scales at which these issuescan be addressed. Producing the ecological knowledge needed foran ecosystem approach is a substantial challenge. Marine sciencesis commonly fragmented, with separate research groups studyingeither open sea or coastal ecosystems, or pelagic or benthic foodwebs, and where fisheries scientists have a different focus thanmarine ecologists. An integration of this scientific knowledgefrom the Baltic Sea is presented below.

2.1. The Baltic Sea ecosystems

The Baltic Sea in its present state is a young (�4000 years)brackish-water ecosystem that consists of a number of topologi-cally defined sub-basins, here defined as the Sound (S: thewestern Baltic transition zone to the North Sea), the Central Baltic(BP) deep basins, the shallower Gulfs of Riga (GR) and Finland (GF)and in the north the Bothnian Sea (BS) and the Bothnian Bay (BB),(Fig. 1). These are subject to a gradient in temperature andsalinity, decreasing from south to north, which primarily resultsin decreasing biodiversity with increasing latitude [28].

The Central Baltic Sea includes the Bornholm and Gotland Basinsand the Gdansk Deep (Fig. 1), the main spawning areas for cod (Gadus

morhua) and sprat (Sprattus sprattus), and the main feeding groundsfor herring (Clupea harengus). The BP is one of the most thoroughlyinvestigated systems when it comes to understanding long-term foodweb dynamics and change, especially for these commerciallyimportant fish species [29–31]. The waters of the Gulfs of Bothnia(BB and BS), Finland (GF) and Riga (GR), (Fig. 1) are partially separatedfrom the BP, and these ecosystems have characteristic patterns anddynamics. However, all basins share common responses to some,mainly climate-related, external driver [32]. Also, many marinespecies of the BP typically migrate into the Gulfs seasonally (e.g.,herring), or extend their distributions into these areas during periodsof high abundance (e.g., cod and sprat; [33]).

The coastal ecosystems are typically structurally complex andlocally variable in comparison to the open sea, and there are alsolarge regional differences. Archipelago areas typically dominatethe northern coastal areas, bays and sounds dominate thesouthwestern coasts, where sandy beaches make up large partsof the southern Baltic Sea, forming an almost linear coastline [34]with extended lagoon systems (Oder, Vistula and Curonianlagoon). Individual coastal ecosystems are often defined by theirhabitat structure, as habitat type has a strong influence on speciesdiversity and distribution. Habitat structure fundamentallydepends on topography, which affects local depth patterns(influencing light regimes), and the level to which an area isexposed to wave action (influencing the bottom substrate). Inaddition, terrestrial influences are of high importance, especiallythe distance to freshwater outflows, land use dynamics (influen-cing diffuse nutrient loads) and coastal constructions.

2.2. Goods and services from the Baltic Sea ecosystems

The 84 million people inhabiting the drainage area [35]derive a number of goods and services from the Baltic Sea.

Fig. 1. The Baltic Sea and its drainage area, including population densities, depth

profiles and features mentioned in the text.

H. Osterblom et al. / Marine Policy 34 (2010) 1290–12991292

One important good is the production of fish for humanconsumption and for industrial purposes. The area also has ahigh recreational value (tourism, recreational fisheries, boating)and several of these services depend on water quality and thestatus of fish stocks. In addition, there are historical and culturalvalues associated with resource uses. The Baltic Sea is alsointensively used for transportation, and was designated as aParticularly Sensitive Sea Area (PSSA) by IMO in 2005 (demandingthe provision of safe traffic routes). Other uses include extractionof e.g., gravel and increasing interest in the establishment ofoffshore wind- and wave energy farms. Human use of theseecosystem goods and services affect the ability of the ecosystemto provide these functions. For example, maritime traffic hasresulted in an increased load of non-native species through ballastwater discharge [36]. Overfishing of cod has not only decreasedthe productivity of this stock [30], thereby affecting bothcommercial and recreational use, but has altered overall fishproductivity and food web dynamics [37]. A decline in fish stocksis also seen to potentially increase symptoms of eutrophication incoastal [38] and open sea [39] systems, calling for an integratedview on ecosystem management.

2.3. Large-scale changes in ecosystem structure and function

The Baltic Sea ecosystems underwent ecological regime shifts2

[40]—in the late 1980s [37,41], both in the open sea, as well as insome coastal areas [32,41]. In the Central Baltic Sea the food web

2 Here defined as a structural change in the ecosystem across multiple trophic

levels over large geographical scales.

changed from a cod- to a sprat-dominated state [39]. Climate-induced hydrographic changes have been identified as the main –but not sole – cause of the Central Baltic Sea ecosystem regimeshift [37]. In large ecosystems, multiple drivers influenceecological regime shifts [42,43]. In the case of the Central Baltic,overfishing decreased the resilience of the cod stock and made itvulnerable to changes in hydrographic conditions [44]. A lack ofinflow of saline water from the North Sea, combined withanthropogenic eutrophication, resulted in oxygen deficiency inthe deepwater layers where cod eggs are neutrally buoyant[45,46], leading to increased cod egg mortality. Furthermore,predator–prey feedback loops have been identified, with a highsprat stock exerting high predation pressure on cod eggs andlarval food [47]. Overfishing as well as eutrophication has beenidentified as additional drivers of regime shifts in the other sub-basins, along with the predominant effect of altered climateforcing [48]. Another example of the effect of multiple, anthro-pogenic and climatic drivers is the altered benthic community inthe Sound. A change in dominance from filter-feeding molluscs topolychaetes was observed [49] under increasing temperatures,decreasing nutrient loads and primary production [50,51]. Thesechanges coincided with the introduction and spread of the non-native polychaete Marenzelleria viridis (a common hitch-hiker ofballast water) to the Baltic Sea [52] and the Sound [53].

The observed regime shifts with large-scale changes inecosystem structure have substantially affected ecosystem func-tion. Following the collapse of the cod stock in the Central BalticSea, sprat was released from predation [39,53–55] and incombination with temperature-driven high recruitment successand increased availability of the warm-water copepod speciesAcartia spp., the sprat stock rose to unprecedented levels [55].These changes appear to have altered the productivity andregulation of zooplankton in the food web [56] as well aspotentially influenced the phytoplankton biomass in summer[39]. The zooplankton community is regulated by climate at lowersprat abundance but top-down controlled by sprat predation athigh sprat stock levels [56]. Thus, the ecological regime shift(influenced both by climate and overfishing) and the trophiccascade that followed changed the regulation of the zooplanktoncommunity from being bottom-up (climate) controlled, to top-down (predation) regulated, with implications also for clupeidgrowth structure [57]. This emphasizes the importance of top-predators in maintaining ecosystem functioning. Furthermore, thedual response of zooplankton to climate or predation by spratpotentially has implications for both cod and ecosystem recovery[39]. In contrast, in the Sound, where a regime shift has also beenobserved [49], no signs of cod collapse or trophic cascades werefound [41]. Whereas atmospheric forcing and the immediateeffects are similar, there is a major difference in anthropogenicforcing, as trawl fishing has been banned in the Sound since 1932[58]. Thus, in the absence of overfishing, ecosystems may responddifferently both to external forcing and to the ecological regimeshifts that may follow.

2.4. Ecological resilience—how serious are the ecosystems shifts?

The ecosystem regime shifts in the Baltic Sea sub-basins haveresulted from multiple stressors: hydro-climatic changes com-bined with overfishing and/or nutrient loading. Thus, althoughecosystems can, potentially, withstand single pressures, multipleanthropogenic stressors acting jointly have a fundamental impacton ecosystem structure and function [42]. The capacity of anecosystem to persist in the face of change and to reorganize after adisturbance depends on its resilience [1]. The differences in theregime shifts that occurred in the Central Baltic Sea (where a

H. Osterblom et al. / Marine Policy 34 (2010) 1290–1299 1293

trophic cascade significantly altered ecosystem structure) and inthe Sound (with no indications of a trophic cascade), suggests thatsingle pressures (overfishing of cod) can erode ecosystemresilience, thereby making it more vulnerable to variations inclimate [59,60].

A crucial question is whether the different ecological regimesdescribed above represent true alternative stable states. If this is thecase it would imply that restoring the ecosystem to a more desiredstate following a regime shift could involve drastic and expensiveinterventions [61,62]—if at all possible. For the Central Baltic Sea ithas been shown that the biotic part of the ecosystem remained inthe new regime, while the external forcing variables (hydro-climaticvariables, nutrients and fishing pressure) have returned to the statebefore the regime shift [37]. This indicates that a new state in thefood web can have been stabilized by feedback loops (hysteresismechanisms) [61]; see [28,29,56,57,61–64] for examples of poten-tial such mechanisms in the Baltic Sea.

2.5. Ecosystem linkages

The active movement of species provides a link betweenecosystems by the transport of energy, predation pressure,competition pressure and genetic variation, but also potentiallyfor the transport of toxins. For example, salmon migrates fromfreshwater rivers in northern Sweden to the open sea in thesouthern Baltic, and back again. Herring and sticklebacks migratebetween feeding areas in the open sea to spawning habitats incoastal areas. Fish as well as zooplankton perform vertical dailymigrations, linking benthic and pelagic food webs in the open seaecosystem.

Water masses containing different levels of salinity, tempera-ture and oxygen, but also different levels of nutrients, toxins andliving planktonic organisms (including drifting fish larvae) aretransported along main currents, and between open sea andcoastal areas. Water exchange between sub-basins and betweencoastal and open areas depends on the water balance of theindividual basin and on topography. The level of water exchangeamong sub-basins drives the distribution of nutrients as influ-enced by external and internal nutrient loading, and alsodetermines the effectiveness of nutrient load reduction. Sub-basins closer to the input of, comparably, nutrient poor North Seawater and with shorter water residence times, generally respondfaster to nutrient load reductions [65]. The central part of theBaltic Sea has a water residence time of several years and theaccumulated nutrient pools stored in the deep layers andsediments trigger internal nutrient loading (especially P) thatby far exceeds the external nutrient load by rivers [66].Nutrient transformations within coastal areas also depend largelyon water residence time [67]. Systems with little water exchangeto open areas are most sensitive to local eutrophication, but atthe same time show the fastest response to reduced nutrientinput [68,69].

Through run-off, diffuse or via inlets, water movement alsolinks coastal marine waters to freshwater ecosystems, as well asto land use affecting run-off. With the exception of relativelypristine estuaries in the Gulf of Bothnia [70], coastal areas derivethe majority of their nutrient budget from terrestrial sources andconsequently have higher nutrient concentrations than adjacentopen sea areas [69,71,72]. The southern Baltic Sea coastal areasreceive the highest nutrient load from rivers, and point sourcesrelated to land use and population patterns [71], which isreflected in a strong impact from freshwater run-off close to thelarge river mouths of the southern and southeastern Baltic [73].Except for the Vistula, the major eutrophicated rivers of the BalticSea flow into sheltered lagoons or gulfs (Fig. 1). The lagoon

systems along the southern and southeastern Baltic Sea shore; theOder lagoon (Oder), Vistula lagoon (Pregolia) and the Curonianlagoon (Nemunas), but also the Gulf of Riga (Daugava) are mostaffected by eutrophication [74]. However, once the river watershave passed these lagoons and gulfs they are rapidly dispersedalong the southern coasts. These open coastlines have a shortwater residence time compared to the sheltered bays andarchipelago areas that are more common in the northern areas[75]. The larger water exchange from these southern coastalareas to the open sea link the large nutrient input to these coastalareas to the Central Baltic Sea.

2.6. Management challenges identified from our ecological

understanding

The review above indicates that many ecological sub-systems ofthe Baltic Sea have exhibited non-linear thresholds, and that thecommon driver of these is climate. It further suggests that multipledrivers – climate and overfishing and/or eutrophication – havepotentially eroded the resilience of the systems and hence thecapacity to respond to changes in climate. These potentiallysynergistic drivers of regime shifts differ between sub-basins,indicating that sub-basin dynamics (coastal-offshore interactions)may be stronger than inter-basin dynamics. This identifies anumber of challenges for governance.

Shifts to alternative regimes that may be stabilized byfeedback mechanisms (a low cod state) underlines the importanceof an ecosystem approach. Multiple drivers at several scales causeproblems in defining appropriate scales for management anddefining responsibilities (and actions) for achieving objectives.Management at small geographic scales may result in fastecosystem response where there are weak linkages to adjacentecosystems. For sub-basin scales however, the relative impor-tance of within and between sub-basin dynamics will determinethe appropriate scales of management.

3. Current governance of the Baltic Sea

As in many areas, environmental policies in the Baltic Searegion have been developed on a sector-by-sector basis (environ-mental, agricultural, fisheries). Since the 1970s the HelsinkiCommission (HELCOM) has been the main forum for internationalenvironmental cooperation (all riparian states are members) [76].The ecosystem approach has, since 2003, been the acceptedframework for HELCOM [77]. The organization has however, notbeen able to reduce the negative effects of eutrophication (aprioritized area) despite almost three decades of politicaldeclarations [78]. Implementation of agreed measures is ham-pered by the voluntary nature of HELCOM cooperation, and thelack of sanctioning mechanisms. EU legislation and policies, incontrast, have strong sanctioning mechanisms and is increasinglyencompassing marine related issues.

3.1. Large-scale changes in governance structure and function

The Helcom Baltic Sea Action Plan (BSAP) [79] addressbiodiversity conservation, hazardous substances and shipping,and has contributed to an ecosystem approach for reducingeutrophication. Total allowable nutrient inputs (critical loadsdefined to achieve ecological targets) have been calculated andspecific country-wise nutrient reduction targets have beenallocated. In addition to this recent change in regional governanceof eutrophication-related issues, there has been a range of recentpolicy developments, which supports a transition to an ecosystem

H. Osterblom et al. / Marine Policy 34 (2010) 1290–12991294

approach. The accession of the Baltic States (Estonia, Latvia andLithuania) to the European Union in 2004, has led to theapplication of a number of top-down changes in governanceframeworks, enabling regional institutional changes. These in-clude the implementation of the habitat [80] and bird [81]directives which influence spatial and species protection mea-sures [82,83]. Importantly, the EU Water Framework Directive[84] has changed the geographical scales at which inland andcoastal waters are managed (Table 2), which provides anopportunity to shift from conventional to more adaptive watermanagement and governance approaches [85].

The European Common Fisheries Policy (CFP) reform in 2003put a larger focus on the ecosystem approach and regional seasand has led to an increased dialogue between actors in RegionalAdvisory Councils (RACs) [86]. Although this change may berather insignificant from a perspective of influencing policyoutcomes [87], it may constitute an important step towards moreregional based co-management of fisheries, emphasized in thecurrent CFP reform process (2009–2012) [88].

The EU Marine Strategy Framework Directive (MSD) [89]constitutes a new framework for the governance of marineresources throughout Europe and could provide a novel approachfor regional social–ecological innovation. At an even higherpolitical level, the EU commission is currently developing anIntegrated Maritime Policy (in which the Marine StrategyDirective is to form the environmental component) addressing awider set of issues such as globalization and competitiveness,climate change, the marine environment, maritime safety, energysecurity and sustainability [90]. These two policies may providemechanisms for integration between water quality, agricultural,fisheries and other policies areas [91]. Maritime spatial planning[92] is intended to be a key instrument for the IntegratedMaritime Policy. These institutional changes result in newperspectives on the scales and basis for management. Ecologicallydefined space (i.e., watersheds or large marine ecosystems, ratherthan administrative regions) is increasingly becoming the startingpoint for European marine-related policies. The conservation ofecosystem structure, function and resilience is becoming a moreclearly articulated priority. However, achieving this in practiceoften means dealing with conflicting interests.

3.2. The implementation gap

Changing policies often start with a change in mental model[1], i.e., the perception of ‘‘how the system works’’. Changes inpublic perception of the Baltic Sea, from a linear, reversible

Table 2Large-scale changes in the governance framework.

Policy Scale Targets

The Baltic Sea Action

Plan 2007

Baltic Sea Reach defined targets

for total nitrogen (N)

and phosphorous (P)

emissions

EU Water

Framework

Directive 2000

National water

districts (inland and

coastal waters)

Reach ‘‘Good

ecological status’’ in

national water district

The 2003 Common

Fisheries Policy

reform

The EU Marine

Strategy Directive

2007

Reach ‘‘Good

environmental status’’

system that is easy to repair, to a system that potentially havegone through regime shifts, resulting in algal blooms [93] anddepleted fish stocks, has likely contributed to the political will insome countries to invest political capital in the Baltic Sea.However, cultural diversity can limit the prospects for findingshared interests and understanding [11] and the environmentalsituation in the Baltic Sea is still much of a non-issue in severalcountries. The countries introducing the EU Water FrameworkDirective tackled this practical implementation problem with aCommon Implementation Strategy (CIS). The CIS working groups,among other, coordinate the intercalibration of ecological statusassessment, leading to an EU Council decision on mandatorywater quality class boundaries for member states [94]. Suchefforts to create shared understanding of processes and problemsare key for developing coherent implementation.

A shared mental model at the political level is crucial forsuccessful implementation of an ecosystem-based approach.Ministers of environment in HELCOM countries took a radicaldecision when agreeing on the nutrient reduction scheme withinthe HELCOM BSAP, despite differences in perception of theproblems and the investments necessary to address them. Also,the Council of Fisheries Ministers in the EU took a historicaldecision in October 2008, when deciding on quotas for cod in linewith recommendations from the International Council for theExploration of the Seas (ICES), for the first time in over a decade.Despite national differences, shared perception of the need forpolitical priority of these issues may be developing.

Making the ecosystem approach operational requires leader-ship and communication skills. Trust building, sense making, thelinking of key individuals, partnerships between actor and thedevelopment of shared vision can mobilize momentum for change[2,95–97]. Leadership means dealing with conflict, e.g., relating tocompetition for space (transport, fishing, offshore energy) [82].Conflict between stakeholders may develop at several levels:between upstream and downstream water users, or betweendifferent categories of fishermen (small- or large scale commercialor recreational fishing). Dealing with conflict needs to take placeat all geographical levels.

Olsen [98] has developed a framework for evaluating progresstowards the implementation of the ecosystem approach, byclassifying four orders of outcome. The large-scale changes inthe governance framework described here primarily representexamples of enabling conditions, or 1st order outcomes. However,relatively limited progress is currently being accomplished at theBaltic Sea level towards 2nd order outcomes, or changes inbehavior and 3rd order outcomes, or harvest, where some socialand environmental quality is maintained, restored or improved

Geographical scope Institutional change Mandate

Scientific targets

defined by ecosystem

model at Baltic Sea and

sub-basin levels

Voluntary agreement

- no sanctioning

mechanisms

Geographical

boundaries define the

scale of management

Water governing

authorities established

Legally binding

Regional Advisory

Council (including

fisheries and

environmental interest

groups) established

Regional Advisory

Council are

consulted by the

Commission

Emphasis on

ecosystem based

approach

Address all human

activities, including

policy coherence

Legally binding

H. Osterblom et al. / Marine Policy 34 (2010) 1290–1299 1295

[98]. The 4th order outcome is described as Sustainable Coastaldevelopment, a more distant goal for the region. However,interesting progress encompassing both 2nd and 3rd orderoutcomes can be observed at local levels (see below). In complexsystems where diverse stakeholders interact, users will attachdifferent priorities to social and environmental objectives. Thelevel of scientific consensus about ecological dynamics anddrivers affecting the system will also influence the definition ofe.g., 4th order outcomes.

3.3. Small scale pilot initiatives create implementation capacity

Reaching targets for marine policies requires adequate resourcesfor implementation and empowerment of stakeholders. A range ofpilot initiatives (Table 3) has been launched to create new capacityto reach objectives in different policy areas. Swedish nationalenvironmental quality objectives related to eutrophication led to theestablishment of the Focus on nutrients project (Table 3, www.greppa.nu), which has created an environment for mutual (social)learning between farmers, agricultural scientists and managers. Ithas also stimulated an interest in adjacent countries (S. Olofsson,Swedish Board of Agriculture, pers.com). Another example ofbottom-up driven governance change can be illustrated by actionstaken by local resource users as a response to national (Swedish)plans to designate a locally important shrimp fishing area as anational park in 1999. The national plans led to the establishment ofa fisheries co-management area [99] and this change at the locallevel stimulated the Swedish Ministry of Agriculture in 2004 toinitiate six pilot initiatives for local co-management of fisheriesalong the Swedish coast [100]. Another pilot initiative forecosystem-based management is the Kristianstads Vattenrike. Theregion in southern Sweden used to be a water logged problem areafor local resource managers. However, it generates a variety ofecosystem services and in the late 1980s and early 1990s themanagement changed through local initiatives, e.g., restoration ofwetlands [96]. Drawing from insights of promising local pilotinitiatives, the Swedish government has recently initiated a numberof pilot projects for local-regional implementation of the ecosystemapproach in the coastal zone (Table 3).

Several of the existing national pilot initiatives were initiatedas a response to crises (problems with reaching national targets,risk of losing important fishing grounds, flooding) and illustrateinstances where: (1) the scope of management has widened

Table 3Recent (Swedish) pilot initiatives in the Baltic Sea area.

Pilot initiative Policy Scale Process

Focus on nutrients

2001-ongoing

CAP,

WFD

Individual

farms

Environmental informa

counseling to reduce lo

of nutrient to air and w

Fisheries Co-management

2004–2007

CFP Local Establish multi-stakeho

collaboration for increa

local autonomy

Kristianstads Vattenrike

(Biosphere reserve)

1989-ongoing

CAP,

WFD

Local Establishment of a boun

organization

Plan fish research project

2008–2013

CFP National/

local

Investigate the role of

zooplanktivores on

ecosystem interactions

Coastal zone planning 2008 MSD Regional Local test of ecosystem

management

from a particular issue to a broader set of issues across scales,(2) management has expanded from individual actors to groups ofactors, (3) knowledge of ecosystem dynamics has developedas a collaborative effort and become part of the organizational andinstitutional structures at multiple levels and (4) social networkshave developed to connect institutions and organizations atmultiple levels to facilitate information flows, identify knowledgegaps, and create nodes of expertise for flexible and collaborativemanagement. These developments have improved the capacityto deal with uncertainty and surprise [96]. Innovative manage-ment experiments are processes which can create novel capacityto deal with ecosystem dynamics and these pilot initiativesare important arenas for innovation, social learning andecosystem based problem solving across sectors and stakeholdergroups [96].

3.4. Towards successful regional implementation

Multiple drivers influence ecosystem dynamics, which differbetween sub-basins. The multitude of drivers impacting change(e.g., nutrients from land and offshore fisheries) illustrates theneed for integration across the land–water interface and betweennational and international governance. Improvement of localconditions can be achieved by local actions (coastal wetlandprotection can improve coastal water quality and stationary fishstocks can be managed locally). Policy change at European andnational levels influence management practices and incentives atnational and local levels. Local initiatives can also create novelcapacity for policy and practice. Successful implementation of theecosystem approach builds on the interactions between science,policy and practice, from local to European levels. Top-downenabling legislation and bottom-up practices can lead to adiffusion of innovation that affect practices at adjacent scales(Fig. 2). The international development of the Baltic Nest modelcreated led to policy change at the regional (Baltic Sea) level(BSAP), impacting national policy and practice (Fig. 2a). Nationalimplementation will impact local and eventually regional levels(a diffusion of innovation, as illustrated by the ovals along thepractice-axis). Science-based changes in agricultural practices toreduce nutrients (Fig. 2b) are in part driven by projects at thenational level (Focus on nutrients). This project has closefeedbacks to science and similar approaches may emerge inother countries. At the same time, policy change at the European

Outcome Partners

tion,

sses

ater

Reduced nutrients and

greenhouse gas emission

Swedish Board of Agriculture, local

authorities, Federation of Swedish

Farmers, etc.

lder

sed

Exchange of knowledge,

conflict mitigation,

development of shared

visions/goals

Commercial fishermen, NGOs,

sports fishermen

dary Integrated management of

ecosystem services

International organizations,

national, regional and local

authorities, corporations,

researchers, non-profit

associations, farmers, landowners

N.A. Swedish Board of fisheries,

Swedish Environmental Protection

Agency (SEPA), International and

National Universities, Commercial

Fishermen Organizations,

environmental NGOs

based N.A. Local stakeholders, coordinated by

SEPA

Fig. 2. Diffusion of innovation and co-evolution between science, policy and practice, when addressing (a) offshore eutrophication, (b) freshwater quality, (c) fisheries

management and (d) integrated marine management. Top-down enabling legislation in combination with science and bottom-up driven initiatives is stimulating the

movement towards a practical implementation of the ecosystem approach.

H. Osterblom et al. / Marine Policy 34 (2010) 1290–12991296

level (WFD) is changing national policies as well as national andlocal practices. The combination of local, national and increasinglyregional changes in practices to address eutrophication areinfluencing the practice-axis at all scales (Fig. 2b). Local pilotinitiatives in fisheries co-management (Fig. 2c) stimulatednational policy change, facilitating additional local initiatives.The CFP reform is influencing stakeholder dialogue at the regionallevel (Baltic Sea Regional Advisory Council). Here, practice atthe local levels is relatively unconnected to practice at theregional level. European policies are increasingly changingregional and national practices through the Marine StrategyFramework Directive and local pilot initiatives (e.g., KristianstadsVattenrike) are stimulated along the national coast of Sweden(Fig. 2d). Potentially, national research programs (Table 3) andregional scientific insights (the existence of regime shifts in therespective sub-basins as reviewed here), can impact on nationaland regional practices, respectively. As in Fig. 2c, there is a lack ofdiffusion of innovation (Fig. 2d) between local and regional levels,which potentially could be stimulated by an increased focus onsub-basin scales.

Changes in policy and practice, with a close link to science,have been evident for the management of water quality (Fig. 2aand b), leading to a change in management regime at allgeographical levels for this large marine ecosystem. Localpractices and European policy reform is stimulating novelapproaches in fisheries governance—although less coherent thanchanges in water governance. These initiatives have primarilybeen stimulated by innovation in practice, rather than by science

(Fig. 2c). Policies for integrated marine management are beingdeveloped at European and national levels. Practical implementa-tion is currently in an early phase and could be further stimulatedby science (Fig. 2d).

The described ecological regime shifts and associated drivershave been identified at sub-basin level, where few of the science–policy–practice interactions are operating. The described govern-ance changes have primarily occurred at local, national and theBaltic Sea scales. To our knowledge, few examples exist ofbilateral or multilateral cooperation around water quality, fish-eries or integrated marine management in the region, focusing onthis scale. Potentially, there is a need for developing governanceregimes with a capacity for addressing the dynamics at this scale,with a focus from watershed to open sea basins interactions. Suchemphasis would likely contribute to reducing existing misfitsbetween ecosystem dynamics and governance structures [5].

A stronger focus on the sub-basin level would allow forenhanced capacity to manage conflict between stakeholders anddeal with sub-basin specific multiple drivers of ecological regimeshifts. The clarity of a regional management framework (e.g.,common methodology and access to data for spatial planning)coupled with a stronger focus on sub-basin monitoring, research,management and adaptation, could contribute with the flexibilitysuggested necessary for robust governance systems [101].This implies that any management body at the sub-basin levelwould maintain a close dialogue with regional management,scientific and stakeholder bodies (such as HELCOM, ICES and theRegional Advisory Council for fisheries, respectively). At this scale,

H. Osterblom et al. / Marine Policy 34 (2010) 1290–1299 1297

co-ordination is likely more feasible than on the regional level. Inaddition, it would also enable bilateral and multilateral coopera-tion across national borders and possibly spur social innovationsfor improved management.

Transaction costs are likely initially high for such formalizedmulti-level governance structures at sub-basin levels. Anyinitiative would build on existing water authorities and relatedbodies. Existing evidence suggests that capacity for implementa-tion can increase substantially, once shared vision, action plans(in line with over-arching goals for the region) and monitoringstrategies are in place [102,103]. However, successful design ofimplementation plans requires ambitious public consultationsand social/economic impact assessments to ensure accountabilityand legitimacy [26,104]. This highlights the dilemma of definingtargets for implementing an ecosystem approach that on the onehand is inclusive of, and acceptable by stakeholders, and on theother hand takes advantage of political windows of opportunity.

4. Is there a role for science to facilitate a transition?

Science has a fundamental role in Baltic Sea policy develop-ment. The NEST decision support system (www.balticnest.org) isan ecosystem model and boundary object [105] primarilydesigned to understand the dynamics and effects of nutrients inthe Baltic Sea (including its drainage area). The system wascritical for reaching political consensus about reductions ofnutrients, aimed at reaching desired environmental status (asdefined by HELCOM experts). The NEST model was used todevelop an allocation scheme for needed actions, perceived as fairby all governments (F. Wulff, Baltic Nest Institute, pers. com.).This unique method of using a decision support system towardsadaptive management shares similarities with the RAINS modelused for reducing acidifying substances [106]. The development ofthe NEST model and recent integrated assessments of Baltic Seafood web dynamics (reviewed above) would have been impos-sible without access to long-term monitoring data. By using allavailable data the ICES/HELCOM Working Group on IntegratedAssessment (WGIAB) could show that the shift in fish communitycomposition was embedded in a larger restructuring of theecosystem at all trophic levels [37,49].

Ecosystem based management is information- and thusmonitoring-intensive, because the interpretation, and responsesto ecosystem feedback at multiple scales require knowledge froma range of different sources [4,107–111]. There is a need also forthe social sciences, economics, and interdisciplinary work, inorder to be able to evaluate the effectiveness of managementoptions or the role of different incentives, as well as scientificobservation of management change processes [26]. The potentialfor international scientific collaboration in the region is increas-ing, with recently initiated joint regional research funding (www.bonusportal.org). Importantly, ICES, providing scientific advice forfish quotas, is currently in a reorganization phase. Consequently,its advisory bodies are changing from giving advice on fisheriesmanagement, marine environment and ecosystems [112], to amore integrated advice on the ecosystem and its dynamics.

This review indicates that more attention should possibly befocused on the sub-basin level. Changes in the scale of manage-ment could imply changes for the organization and role of thescientific community. Science could play an important role infacilitating routines for assessments and stakeholder dialogues atthe sub-basin level, thereby creating collaborative learning plat-forms and boundary organizations [105]. Important exercisescould include defining use interaction matrices, describingpotential conflict between sectors [17] and identification of thenegative impacts of human use on ecosystems, as well as impacts

of ecosystem degradation on the production of desired ecosystemgoods and services [17]. The process of defining such matricesrequires that scientific expertise, and stakeholder and localknowledge is used constructively. Social–ecological inventories,where ecological inventories are complemented by inventories oflocal stakeholder groups, interests and management initiatives[113], are necessary for understanding context and potential forcollaborative learning and ecosystem based management. Scien-tific networks could initiate stakeholder inclusive, science-basedprocesses, where emerging learning platforms would contributewith knowledge for regional ecosystem based management. Byplaying the role of honest broker, one developing role of sciencecould be to bridge critical gaps between science, practice andpolicy, as well as between stakeholders. Following these processeswithin the social sciences could lead to important scientificinsights about governance of social–ecological systems. Ideally,analogous learning platforms should be set up for the regional(Baltic Sea) level in order to better understand cross-scaleinteractions in governance structures.

5. Conclusions

The ecosystem approach to the management of marineresources is commonly called for, but application differs sub-stantially between regions. Interesting and innovative approachesare however apparent e.g., in the Barents Sea region [114], theGreat Barrier Reef [26], the Puget Sound [25] and in the Antarctic[115]. This study illustrates that even a comparatively simplebiological marine system, with low species diversity is complex tomanage, resulting from ecological regime shifts, cross-scaleecological interactions and social–ecological dynamics. The BalticSea is substantially shaped by human activities, some of whichhave to be managed in an international context, others that can beaddressed at local and sub-basin levels. Interesting innovations toaddress social–ecological dynamics at different spatial scales areemerging, some of which have spread to other spatial scales orsectors. This diffusion of innovation has been made possiblethrough interactions between science, policy and practice. Thepotential for local innovation and the spread of ideas isemphasized as a key tool for implementing the ecosystem-basedapproach. Although recent political actions to reduce overfishingand eutrophication, as well as expected future initiatives formarine spatial planning are important, these actions are likelyrelatively inefficient without stakeholder consultation, engage-ment and participation across interest groups. Emphasis on multi-level governance structures that provides space for experimentingand spread of social innovations at local and regional scales canprovide key elements for stimulating an adaptive capacity fordealing with this dynamic ecosystem and the services generated.One way forward could be to establish a number of stakeholderinclusive collaborative learning platforms at the sub-basinscale—with a clear mandate and with the aim to address spatiallyrelevant dynamics.

Acknowledgements

We appreciate the assistance from E. Smedberg and R. Kautskyfor the production of Figs. 1 and 2, respectively. This study was inpart funded by the EC FP7 project: Knowledge-based SustainableManagement for Europe’s Seas (KnowSeas), Grant agreementno. 226675.

H. Osterblom et al. / Marine Policy 34 (2010) 1290–12991298

References

[1] Walker B, Holling CS, Carpenter SR, Kinzig A. Resilience, adaptability andtransformability in social–ecological systems. Ecology and Society 2004;9.

[2] Folke C, Hahn T, Olsson P, Norberg J. Adaptive governance of social–ecological systems. Annual Review of Environment and Resources2005;30:441–73.

[3] Dietz T, Ostrom E, Stern PC. The struggle to govern the commons. Science2003;302:1907–12.

[4] Olsson P, Folke C, Berkes F. Adaptive comanagement for building resiliencein social–ecological systems. Environmental Management 2004;34:75–90.

[5] Galaz V, Olsson P, Folke C, Hahn T. The problem of fit between ecosystemsand governance systems—insights and emerging challenges. In: Young O,editor. Earth system governance: a research agenda. MIT Press; 2008.

[6] Ostrom V. Nonhierarchical approaches to the organization of public activity.Annals of the American Academy of Political and Social Science1983;466:135–47.

[7] Hooghe L, Marks G. Unraveling the central state, but how? Types of multi-level governance American Political Science Review 2003;97:233–43.

[8] van Kersbergen K, van Waarden F. ‘Governance’ as a bridge betweendisciplines: cross-disciplinary inspiration regarding shifts in governanceand problems of governability, accountability and legitimacy. EuropeanJournal of Political Research 2004;43:143–71.

[9] Baland JM, Platteau JP. Halting degredation of natural resources: is there arole for rural communities? Oxford: Clarendon Press; 1996.

[10] Holling CS, Meffe GK. Command and control and the pathology of naturalresource management. Conservation Biology 1996;10:328–37.

[11] Ostrom E, Burger J, Field CB, Norgaard RB, Policansky D. Sustainability—

revisiting the commons: local lessons, global challenges. Science1999;284:278–82.

[12] CBD. Report of the Workshop on the Ecosystem Approach, Lilongwe,Malawi; 1998.

[13] Garcia SM, Zerbi A, Aliaume C, Do Chi T, Lasserre G. The ecosystem approachto fisheries. Issues, terminology, principles, institutional foundations,implementation and outlook. FAO Fisheries Technical Paper. Rome: FAO;2003. p. 71.

[14] UNEP/GPA. Ecosystem-based management: markers for assessing progress.The Hague: UNEP/GPA; 2006.

[15] Hughes TP, Gunderson LH, Folke C, Baird AH, Bellwood D, Berkes F, et al.Adaptive management of the great barrier reef and the Grand Canyon worldheritage areas. Ambio 2007;36:586–92.

[16] Pikitch EK, Santora C, Babcock EA, Bakun A, Bonfil R, Conover DO, et al.Ecosystem-based fishery management. Science 2004;305:346–7.

[17] Juda L, Hennessey T. Governance profiles and the management of the usesof large marine ecosystems. Ocean Development & International Law2001;32:43–69.

[18] Hughes TP, Bellwood DR, Folke C, Steneck RS, Wilson J. New paradigms forsupporting the resilience of marine ecosystems. Trends in Ecology &Evolution 2005;20:380–6.

[19] Rice J. Implementation of the ecosystem approach to fisheries manage-ment—asynchronous co-evolution at the interface between science andpolicy. Marine Ecology Progress Series 2005;275:265–70.

[20] Mahon R, Fanning L, McConney P. A governance perspective on the largemarine ecosystem approach. Marine Policy 2009;33:317–21.

[21] Juda L. Considerations in developing a functional approach to thegovernance of large marine ecosystems. Ocean Development & Interna-tional Law 1999;30:89–125.

[22] Browman HI, Cury PM, Hilborn R, Jennings S, Lotze HK, Mace PM, et al.Theme section: perspectives on ecosystem-based approaches to themanagement of marine resources. Marine Ecology Progress Series2004;274:269–303.

[23] Browman HI, Stergiou KI. Politics and socio-economics of ecosystem-basedmanagement of marine resources. Marine Ecology Progress Series2005;300:241–2.

[24] Levin PS, Fogarty MJ, Murawski SA, Fluharty D. Integrated ecosystemassessments: developing the scientific basis for ecosystem-based manage-ment of the ocean. PLoS Biology 2009;7:e1000014, doi:10.1371/journal.-pbio.

[25] Tallis H, Levin PS, Ruckelshaus M, Lester SE, McLeod KL, Fluharty DL, et al.The many faces of ecosystem-based management: making the process worktoday in real places. Marine Policy 2010;34:340–8.

[26] Olsson P, Folke C, Hughes TP. Navigating the transition to ecosystem-basedmanagement of the Great Barrier Reef, Australia. Proceedings of theNational Academy of Sciences of the United States of America2008;105:9489–94.

[27] Levin PSFM, Murawski SA, Fluharty D. Integrated ecosystem assessments:developing the scientific basis for ecosystem-based management of theocean. PLoS Biology 2009;7:e1000014, doi:10.1371/journal.pbio.

[28] Elmgren R, Hill C. Ecosystem function at low biodiversity—the Balticexample. In: Ormond RFG, Gage JD, Angel MV, editors. Marinebiodiversity pattern and processes. Cambridge: Cambridge University Press;1997. p. 319–36.

[29] Mollmann C, Koster FW. Population dynamics of calanoid copepods and theimplications of their predation by clupeid fish in the Central Baltic Sea.Journal of Plankton Research 2002;24:959–77.

[30] Koster FW, Mollmann C, Hinrichsen H-H, Tomkiewicz J, Wieland K, Kraus G,et al. Baltic cod recruitment—the impact of climate and species interaction.ICES Journal of Marine Science 2005;62:1408–25.

[31] Bagge O, Thurow F, Steffensen E, Bay J. The Baltic cod. Dana 1994;10:1–28.[32] Diekmann R, Mollmann C. Integrated ecosystem assessments of seven Baltic

Sea areas covering the last three decades. ICES cooperative research report305. Copenhagen: ICES; 2010. p. 92.

[33] Aro E. A review of fish migration patterns in the Baltic. Rapports et Proces-verbaux des Reunions Conseil International pour l’Exploration de la Mer1989;190:70–96.

[34] Al-Hamdani Z, Reker J. Towards marine landscapes in the Baltic Sea.BALANCE Interim Report; 2007. p. 116.

[35] Hannerz F, Destouni G. Spatial characterization of the Baltic Sea DrainageBasin and its unmonitored catchments. Ambio 2006;35:214–9.

[36] Almqvist G. Round goby Neogobius melanostomus in the Baltic Sea:invasion biology in practice. Stockholm: Stockholm University; 2008.

[37] Mollmann C, Diekmann R, Muller-Karulis B, Kornilovs G, Plikshs M, Axe P.Reorganization of a large marine ecosystem due to atmospheric andanthropogenic pressure: a discontinuous regime shift in the Central BalticSea. Global Change Biology 2009;15:1377–93.

[38] Eriksson BK, Ljunggren L, Sandstrom A, Johansson G, Mattila J, Rubach A,et al. Declines in predatory fish promote bloom-forming macroalgae.Ecological Applications 2009;19:1975–88.

[39] Casini M, Lovgren J, Hjelm J, Cardinale M, Molinero JC, Kornilovs G. Multi-level trophic cascades in a heavily exploited open marine ecosystem.Proceedings of the Royal Society of London, Series B: Biological Sciences2008;275:1793–801.

[40] Scheffer M, Carpenter S, Foley JA, Folke C, Walker B. Catastrophic shifts inecosystems. Nature 2001;413:591–6.

[41] ICES. Report of the ICES/HELCOM working group on integrated assessmentsof the Baltic Sea (WGIAB). Copenhagen: ICES; 2008. p. 84.

[42] Folke C, Carpenter S, Walker B, Scheffer M, Elmqvist T, Gunderson L, et al.Regime shifts, resilience, and biodiversity in ecosystem management.Annual Review of Ecology Evolution and Systematics 2004;35:557–81.

[43] Daskalov GM, Grishin AN, Rodionov S, Mihneva V. Trophic cascadestriggered by overfishing reveal possible mechanisms of ecosystem regimeshifts. Proceedings of the National Academy of Sciences of the United Statesof America 2007;104:10518–23.

[44] Koster FW, Mollmann C, Hinrichsen HH, Wieland K, Tomkiewicz J, Kraus G,et al. Baltic cod recruitment—the impact of climate variability on keyprocesses. ICES Journal of Marine Science 2005;62:1408–25.

[45] MacKenzie BR, Hinrichsen H-H, Plikshs M, Wieland K, Zezera AS. Quantify-ing environmental heterogeneity: habitat size necessary for successfuldevelopment of cod Gadus morhua eggs in the Baltic Sea. Marine EcologyProgress Series 2000;193:143–56.

[46] Wieland K, Waller U, Schnack D. Development of Baltic cod eggs at differentlevels of temperature and oxygen content. Dana 1994;10:163–77.

[47] Mollmann C, Muller-Karulis B, Kornilovs G, John St. MA. Effects of climateand overfishing on zooplankton dynamics and ecosystem structure: regimeshifts, trophic cascade, and feedback coops in a simple ecosystem. ICESJournal of Marine Science 2008;65:302–10.

[48] Blenckner T. Coherent and basin-specific drivers cause ecological regimeshifts in the Baltic Sea, unpublished results.

[49] Lindegren M, Diekmann R, Mollmann C. Regime shifts, resilience andrecovery of a local cod stock. Marine Marine Ecology Progress Series2010;402:239–53.

[50] Wasmund N, Uhlig S. Phytoplankton trends in the Baltic Sea. ICES Journal ofMarine Science 2003;60:177–86.

[51] Henriksen P. Long-term changes in phytoplankton in the Kattegat, the BeltSea, the Sound and the western Baltic Sea. Journal of Sea Research2009;61:114–23.

[52] Bochert R, Zettler ML, Bochert A. Variation in the reproductive status, larvaloccurrence and recruitment in an estuarine population of Marenzelleriaviridis (Polychaeta: Spionidae). Ophelia 1996;45:127–42.

[53] Stromberg A, Persson P. Den Amerikanske havsborstmasken Marenzelleriaviridis langs Helsingborgkusten 2005. Helsingborg: Miljonamnden iHelsingborg; 2005.

[54] Koster FW, Hinrichsen HH, Schnack St D, John MA, Mackenzie BR,Tomkiewicz J, et al. Recruitment of Baltic cod and sprat stocks:identification of critical life stages and incorporation of environmentalvariability into stock-recruitment relationships. Scientia Marina2003;67:129–54.

[55] Mackenzie BR, Gislason H, Mollmann C, Koster FW. Impact of 21st centuryclimate change on the Baltic Sea fish community and fisheries. GlobalChange Biology 2007;13:1348–67.

[56] Casini M, Hjelm J, Molinero JC, Lovgren J, Cardinale M, Bartolino V, et al.Trophic cascades promote threshold-like shifts in pelagic marine ecosys-tems. Proceedings of the National Academy of Sciences of the United Statesof America 2009;106:197–202.

[57] Casini M, Bartolino V, Molinero JC, Kornilovs G. Linking fisheries, climateand trophic interactions: threshold dynamics drive herring (Clupeaharengus) growth in the central Baltic Sea. Marine Ecology Progress Series,in press, doi:10.3354/meps08592.

[58] Svedang H, Hagberg J, Borjesson P, Svensson A, Vitale F. Bottenfisk iVasterhavet—Fyra studier av bestandens status, utveckling och lekomradenvid den svenska vastkusten. Finfo. Gothenburg: Board of Fisheries; 2004.

H. Osterblom et al. / Marine Policy 34 (2010) 1290–1299 1299

[59] Brander KM. Climate Change and Food Security Special Feature: Global fishproduction and climate change. Proceedings of the National Academy ofSciences of the United States of America 2007;104:19709–14.

[60] Brander KM. Cod recruitment is strongly affected by climate when stockbiomass is low. ICES Journal of Marine Science 2005;62:339–43.

[61] Scheffer M, Carpenter SR. Catastrophic regime shifts in ecosystems: linkingtheory to observation. Trends in Ecology & Evolution 2003;18:648–56.

[62] Suding KN, Gross KL. Houseman GR. Alternative states and positive feedbacksin restoration ecology. Trends in Ecology & Evolution 2004;19:46–53.

[63] Koster FW, Mollmann C. Egg cannibalism in Baltic sprat Sprattus sprattus.Marine Ecology Progress Series 2000;196:269–77.

[64] Van Leeuwen A, De Roos AM, Persson L. How cod shapes its world. Journalof Sea Research 2008;60:89–104.

[65] Savchuk OP, Wulff F. Modeling the Baltic Sea eutrophication in a decisionsupport system. Ambio 2007;36:141–8.

[66] Savchuk OP, Wulff F, Hille S, Humborg C, Pollehne F. The Baltic Sea a centuryago—a reconstruction from model simulations, verified by observations.Journal of Marine Systems 2008;74:485–94.

[67] Wallin M, L. Hakanson. Morphometry and sedimentation as regulatingfactors for nutrient reccling and trophic state in coastal waters. Hydro-biologia 1992;235:33–45.

[68] Carstensen J, Conley DJ, Andersen JH, Aertebjerg G. Coastal eutrophicationand trend reversal: a Danish case study. Limnology and Oceanography2006;51:398–408.

[69] HELCOM. Eutrophication in the Baltic Sea—an integrated thematic assess-ment of the effects of nutrient enrichment in the Baltic Sea Region. BalticSea environment proceedings. Helsinki 2009.

[70] Humborg C, Danielsson A, Sjoberg B, Green M. Nutrient land-sea fluxes inoligothrophic and pristine estuaries of the Gulf of Bothnia, Baltic Sea.Estuarine, Coastal and Shelf Science 2003;56:781–93.

[71] Rahm L, Danielsson A. Spatial heterogeneity of nutrients in the BalticProper, Baltic Sea. Estuarine Coastal and Shelf Science 2007;73:268–78.

[72] Voss M, Emeis KC, Hille S, Neumann T, Dippner JW. Nitrogen cycle of the BalticSea from an isotopic perspective. Global Biogeochemical Cycles 2005:19.

[73] Meier HEM. Modeling the pathways and ages of inflowing salt- andfreshwater in the Baltic Sea. Estuarine, Coastal and Shelf Science2007;74:610–27.

[74] Humborg C, Conley DJ, Rahm L, Wulff F, Cociasu A, Ittekkot V. Siliconretention in river basins: far-reaching effects on biogeochemistry andaquatic food webs in coastal marine environments. Ambio 2000;29:45–50.

[75] Schernewski G, Wielgat MA. Baltic Sea typology according to the EC-WaterFramework Directive: integration of national typologies and the water bodyconcept. Baltic Sea Typology Coastline Reports 2004. p. 1–26.

[76] Selin H, VanDeveer SD. Baltic Sea hazardous substances management:results and challenges. Ambio 2004;33:153–60.

[77] Backer H, Leppanen JM. The HELCOM system of a vision, strategic goals andecological objectives: implementing an ecosystem approach to the manage-ment of human activities in the Baltic Sea. Aquatic Conservation—Marineand Freshwater Ecosystems 2008;18:321–34.

[78] Kremser U. Agriculture within the context of HELCOM’s—Mandate andactivities. Ambio 1997;26:415–7.

[79] HELCOM. Baltic Sea action plan. Helcom ministerial meeting. Krakow:HELCOM; 2007.

[80] EEC. Council Directive 92/43/EEC of 21 May 1992 on the conservation ofnatural habitats and of wild fauna and flora. Official Journal of the EuropeanUnion 1992.

[81] EEC. Council Directive 79/409/EEC of 2 April 1979 on the conservation ofwild birds. Official Journal of the European Union 1979.

[82] Salomon M. Recent European initiatives in marine protection policy:towards lasting protection for Europe’s seas?Environmental Science &Policy 2009;12:359–66.

[83] Juda L. The European union and ocean use management: the marinestrategy and the maritime policy. Ocean Development & International Law2007;38:259–82.

[84] EC. Directive 2000/60/EC of the European Parliament and of the Councilestablishing a framework for the Community action in the field of waterpolicy, Brussels; 2000.

[85] Olsson P, Galaz V. Transitions to adaptive approaches to water managementand governance in Sweden. In: Huitema D, Meijerink S, editors. Waterpolicy entrepreneurs: a research companion to water transitions around theglobe. Northampton: Edward Elgar Publishing, Inc.; 2010. p. 304–24.

[86] Daw T, Gray T. Fisheries science and sustainability in international policy: astudy of failure in the European Union’s Common Fisheries Policy. MarinePolicy 2005;29:189–97.

[87] Gray T, Hatchard J. The 2002 reform of the Common Fisheries Policy’ssystem of governance—rhetoric or reality?Marine Policy 2003;27:545–54.

[88] COM. Green paper: reform of the common fisheries policy. Brussels: TheEuropean Commission; 2009.

[89] EC. Directive 2008/56/EC of the European Parliament and of the Council of17 June 2008 establishing a framework for community action in the field of

marine environmental policy (Marine Strategy Framework Directive),Brussels; 2008.

[90] COM. An Integrated Maritime Policy for the European Union. Communica-tion from the Commission to the European Parliament, the Council, theEuropean Economic and Social Committee and the Committee of theRegions, Brussels; 2007.

[91] Juda L. The European union and ocean use management: the marinestrategy and the maritime policy. Ocean Development and InternationalLaw 2007;38:259–82.

[92] COM. Roadmap for Maritime Spatial Planning: Achieving CommonPrinciples in the EU—Communication from the Commission, Brussels; 2008.

[93] Osterblom H, Hansson S, Larsson U, Hjerne O, Wulff F, Elmgren R, et al.Human-induced trophic cascades and ecological regime shifts in the Balticsea. Ecosystems 2007;10:877–89.

[94] COM. Establishing, pursuant to Directive 2000/60/EC of the EuropeanParliament and of the Council, the values of the Member State monitoringsystem classifications as a result of the intercalibration exercise. OfficialJournal of the European Union 2008.

[95] Berkes F, Colding F, Folke C. Navigating social–ecological systems: buildingresilience for complexity and change. Cambridge: Cambridge UniversityPress; 2003.

[96] Olsson P, Folke C, Hahn T. Social–ecological transformation for ecosystemmanagement: the development of adaptive co-management of a wetlandlandscape in southern Sweden. Ecology and Society 2004;9.

[97] Olsson P, Gunderson LH, Carpenter SR, Ryan P, Lebel L, Folke C, et al.Shooting the rapids: navigating transitions to adaptive governance ofsocial–ecological systems. Ecology and Society 2004;11.

[98] Olsen SB. Frameworks and indicators for assessing progress in integratedcoastal management initiatives. Ocean & Coastal Management2003;46:347–61.

[99] Pıriz L. Hauling home the co-management of coastal fisheries: a study oninstitutional barriers to fishermen’s involvement in the management ofcoastal fisheries on the West Coast of Sweden. Gothenburg: GothenburgUniversity; 2004.

[100] Anonymous. Regional of lokal samforvaltning av fiske Gothenburg: SwedishBoard of Fisheries; 2007.

[101] Duit A, Galaz V. Governance and complexity—emerging issues for govern-ance theory. Governance 2008;21:311–35.

[102] Evans KE, Klinger T. Obstacles to bottom-up implementation of marineecosystem management. Conservation Biology 2008;22:1135–43.

[103] Reed MG. Environmental justice and community-based ecosystem manage-ment. In: Haluza-DeLay R, Agyeman J, Cole P, O’Riley, editors. Our ownvoice: constructions of environmental justice in Canada. Vancouver: UBCPress; 2008.

[104] Lebel L, Anderies JM, Campbell B, Folke C, Hatfield-Dodds S, Hughes TP, et al.Governance and the capacity to manage resilience in regional social–ecological systems. Ecology and Society 2004;11.

[105] Cash DW, Clark WC, Alcock F, Dickson NM, Eckley N, Guston DH, et al.Knowledge systems for sustainable development. Proceedings of theNational Academy of Sciences of the United States of America2003;100:8086–91.

[106] Amann M. Some Lessons from the use of the RAINS model in internationalnegotiations. Springer; 2007.

[107] Gadgil M, Olsson P, Berkes F, Folke C. Exploring the role of local ecologicalknowledge for ecosystem management: three case studies. In: Berkes F,Colding J, Folke C, editors. Navigating social–ecological systems: buildingresilience for complexity and change. Cambridge: Cambridge UniversityPress; 2003.

[108] Carlsson L, Berkes. F. Comanagement across levels of organization: conceptsand methodological implications. Journal of Environmental Management2005;75:65–76.

[109] Berkes F. Cross-scale institutional linkages: perspectives from thebottom up. In: Ostrom E, Dietz T, Dolsak N, Stern P, Stonich S,Weber EU, editors. The drama of the commons. Washington: NationalAcademy Press; 2002. p. 293–321.

[110] Brown K. Integrating conservation and development: a case of institutionalmisfit. Frontiers in Ecology and the Environment 2003;1:479–87.

[111] Folke C. Traditional knowledge in social–ecological systems. Ecology andSociety 2004;9.

[112] Misund OA, Skjoldal HR. Implementing the ecosystem approach: experi-ences from the North Sea, ICES, and the Institute of Marine Research,Norway. Marine Ecology Progress Series 2005;300:260–5.

[113] Schultz L, Folke C, Olsson P. Enhancing ecosystem management throughsocial–ecological inventories: lessons from Kristianstads Vattenrike,Sweden. Environmental Conservation 2007;34:140–52.

[114] Olsen E, Gjøsæter H, Røttingen I, Dommasnes A, Fossum P, Sandberg P. TheNorwegian ecosystem-based management plan for the Barents Sea. ICESJournal of Marine Science 2007;64:599–602.

[115] Murawski SA. Ten myths concerning ecosystem approaches to marineresource management. Marine Policy 2007;31:681–90.