The Future Oceanscgmdesign.fatcow.com/EPBRS/Documents/Reports/Future...mark) (2005): Assessment of...

Publications of the German Advisory Council on Global Change (WBGU)

World in Transition: Fighting Poverty through Environmental Policy. 2004 Report.London: Earthscan © 2005, ISBN 1-85383-883-7.

World in Transition: Towards Sustainable Energy Systems. 2003 Report.London: Earthscan © 2004, ISBN1-85383-882-9.

Renewable Energies for Sustainable Development: Impulses for renewables 2004. Policy Paper 3.Berlin: WBGU © 2004, 24 pages, ISBN 3-936191-06-9.

Climate Protection Strategies for the 21st Century: Kyoto and beyond. Special Report 2003.Berlin: WBGU © 2003, ISBN 3-936191-04-2.

Charging the Use of Global Commons. Special Report 2002.Berlin: WBGU © 2002, ISBN 3-9807589-8-2.

Charges on the Use of the Global Commons. WBGU Policy Paper 2.Berlin: WBGU © 2002, 16 pages, ISBN 3-936191-00-X.

World in Transition: New Structures for Global Environmental Policy. 2000 Report.London: Earthscan © 2001, ISBN 1-85383-852-7.

World in Transition: Conservation and Sustainable Use of the Biosphere. 1999 Report.London: Earthscan © 2001, ISBN 1-85383-802-0.

The Johannesburg Opportunity: Key Elements of a Negotiation Strategy. WBGU Policy Paper 1. Berlin: WBGU © 2001, 20 pages, ISBN 3-9807589-6-6.

World in Transition: Strategies for Managing Global Environmental Risks. 1998 Report.Berlin: Springer © 2000, ISBN 3-540-66743-1.

World in Transition: Environment and Ethics. 1999 Special Report.Website: http://www.wbgu.de/WBGU/wbgu_sn1999_engl.html.

World in Transition: Ways Towards Sustainable Management of Freshwater Resources. Report 1997. Berlin: Springer © 1998, ISBN 3-540-64351-6.

The Accounting of Biological Sinks and Sources Under the Kyoto Protocol - A Step Forwards or Backwards for Global Environmental Protection? Special Report 1998. Bremerhaven: WBGU © 1998.ISBN 3-9806309-1-9.

World in Transition: The Research Challenge. 1996 Report. Berlin: Springer © 1997. ISBN 3-540-61832-5.

World in Transition: Ways Towards Global Environmental Solutions. 1995 Report.Berlin: Springer © 1996. ISBN 3-540-61016-2.

World in Transition: The Threat to Soils. 1994 Report. Bonn: Economica © 1995. ISBN 3-87081-055-6.

World in Transition: Basic Structure of Global People-Environment Interactions. 1993 Report.Bonn: Economica © 1994. ISBN 3-87081-154-4.

All WBGU Reports can be downloaded through the Internet from the website http://www.wbgu.de

ISBN 3-936191-14-X

Special Report

R. SchubertH.-J. SchellnhuberN. BuchmannA. EpineyR. GrießhammerM. KulessaD. MessnerS. RahmstorfJ. Schmid

German Advisory Council on Global Change(WBGU)

WBGU

The Future Oceans – Warming Up, Rising High, Turning Sour


C

M

Y

CM

MY

CY

CMY

K

wbgu_sn2006_deckel_en_2.ai 26.07.2006 13:08:13 Uhrwbgu_sn2006_deckel_en_2.ai 26.07.2006 13:08:13 Uhr

Members of the German Advisory Council on Global Change (WBGU)(as on 1. January 2006)

Prof Dr Renate Schubert (chair), economistHead of the Center for Economic Research at ETH Zurich, Switzerland

Prof Dr Hans Joachim Schellnhuber CBE (vice chair), physicistDirector of the Potsdam Institute for Climate Impact Research (PIK) and visiting professor at Oxford University

Prof Dr Nina Buchmann, ecologistProfessor of Grassland Science, Institute of Plant Sciences at ETH Zurich, Switzerland

Prof Dr Astrid Epiney, lawyerDirector of the Institute for European Law, Université de Fribourg, Switzerland

Dr Rainer Grießhammer, chemical engineerMember of the management board of the Institute for Applied Ecology, Freiburg

Prof Dr Margareta E. Kulessa, economistProfessor of International Economics at the University of Applied Sciences, Mainz

Prof Dr Dirk Messner, political scientistDirector of the German Development Institute, Bonn

Prof Dr Stefan Rahmstorf, physicistProfessor of Physics of the Oceans, Potsdam University and head of the Climate System Department at the Potsdam Institute for Climate Impact Research

Prof Dr Jürgen Schmid, aerospace engineerPresident of the Institute for Solar Energy Technology and Professor at the University of Kassel,Head of the Department for Efficient Energy Conversion

German Advisory Council on Global Change

The Future Oceans –Warming up, Rising High,Turning Sour

Special Report

Berlin 2006

GERMAN ADVISORY COUNCIL ON GLOBAL CHANGEWissenschaftlicher Beirat der Bundesregierung Globale Umweltveränderungen (WBGU)Secretariat Reichpietschufer 60–62, 8th floorD-10785 Berlin

Phone +49 (0) 30 263948 0Fax +49 (0) 30 263948 50E-Mail [email protected] http://www.wbgu.de

Translation: Christopher Hay, Seeheim-Jugenheim, [email protected]

Copy deadline: 23. March 2006

This special report is available through the Internet in German and English

ISBN 3-936191-14-X

Cover: Sleipner gas platform in the North Sea (M. Schulz-Baldes, WBGU), ‘Elephant Foot Glacier’ in theArctic (H. Oerter, AWI Bremerhaven), coral reef (Dan Barbus, Romania), illegal fishing in the Pacific (Aus-tralian Customs Service)

It would not have been possible to produce this spe-cial report without the committed and untiringefforts of the staff of the Council Members and theWBGU Secretariat in Berlin.

The scientific team participating in the work ofWBGU when this report was written included:

Prof Dr Meinhard Schulz-Baldes (WBGU Secre-tary-General), Dr Carsten Loose (WBGU DeputySecretary-General, WBGU Secretariat Berlin), Stef-fen Bauer, MA (Environmental Policy ResearchCentre, Free University of Berlin, since 01.01.2006),Dr Gregor Betz (Potsdam Institute for ClimateImpact Research – PIK, until 30.09.2005), Dipl.-Phys.Gregor Czisch (Institute for Electrical Engineer-ing/Efficient Energy Conversion (IEE-RE), Kassel,until 01.03.2006), Dipl.-Volksw. Oliver Deke(WBGU Secretariat Berlin, from 17.10.2005), Dipl.-Umweltwiss. Tim Hasler (WBGU SecretariatBerlin), Dr Monika Heupel (University of Bremen,until 15.10.2005), Dipl.-Volksw. Kristin Hoffmann(Swiss Federal Institute of Technology Zurich), DrSusanne Kadner (PIK, Potsdam, until 30.04.2006), DrSabina Keller (Swiss Federal Institute of TechnologyZurich), Dipl.-Pol. Lena Kempmann (WBGU Secre-tariat Berlin), Dipl.-Geogr. Andreas Manhart (Insti-tute for Applied Ecology, Freiburg), Dr FranziskaMatthies (University of Copenhagen, until31.10.2005), Dr Nina V. Michaelis (WBGU Secre-tariat Berlin, until 18.11.2005), Dipl.-Volksw. MarkusOhndorf (Swiss Federal Institute of TechnologyZurich), Dr Benno Pilardeaux (WBGU SecretariatBerlin), Dr Martin Scheyli (University of Fribourg,Switzerland), Dr Astrid Schulz (WBGU SecretariatBerlin), Dipl.-Pol. Joachim Schwerd (University ofApplied Sciences, Mainz).

WBGU is also grateful for the important contri-butions and support provided by other members ofthe research community.This special report builds onthe following expert’s studies which were commis-sioned by WBGU:• Prof David Archer (Department of Geophysical

Sciences, University of Chicago) (2006): Destabi-lization of Methane Hydrates: A Risk Analysis.

• Dr Nick Brooks, Prof Dr Robert Nicholls, Prof DrJim Hall (Tyndall Centre for Climate ChangeResearch, University of East Anglia, Norwich,UK) (2006): Sea Level Rise: Coastal Impacts andResponses.

• Dr Keith Brander (International Council for theExploration of the Sea – ICES, Copenhagen, Den-mark) (2005): Assessment of Possible Impacts ofClimate Change on Fisheries.

• Prof Dr Hans-Otto Pörtner (Alfred WegenerInstitute for Polar- and Marine Research, Bremer-haven, Germany) (2005):Auswirkungen von CO2-Eintrag und Temperaturerhöhung auf die marineBiosphäre.

WBGU would also like to thank all those expertswho gave us their opinion on drafts of the specialreport, providing us with invaluable comments andadvice:– Dr Peter G. Brewer (Monterey Bay Aquarium

Research Institute, USA),– Prof Atsushi Ishimatsu (Nagasaki University,

Japan),– Dr James Orr (Laboratoire des Sciences du

Climat et de l’Environnement, Gif-Sur-Yvette,Frankreich),

– Prof Dr Ulf Riebesell (Leibniz Institute of MarineSciences, Kiel, Germany).

In addition, for providing peer reviews or contribu-tions to individual chapters of the report, WBGUalso thanks Dipl.-Phys. Jochen Bard (Institute forSolar Energy Technology – ISET, Kassel), DrMatthias Hofmann (PIK, Potsdam) and Dr CorinneLe Quéré (School of Environmental Sciences, Uni-versity of East Anglia, Norwich, UK).

Sincere thanks also go to the organizers and dis-cussion partners who were involved in the WBGUstudy trip to Norway from 6 to 14 October 2005.Many experts from the fields of government admin-istration, politics and science prepared visits, eventsand presentations for WBGU and made themselvesavailable for discussions. WBGU would like toexpress special thanks to Ambassador RolandMauch and our contact person in the German

Acknowledgements

Embassy in Oslo, Ms Charlotte Schwarzer, for theirsupport in organizing the trip.

WBGU is also indebted to the staff of the follow-ing institutions and companies for the useful discus-sions and conversations we had with them:– SINTEF – Stiftelsen for industriell og teknisk

forskning, Trondheim,– NTNU – Norges teknisk-naturvitenskapelige uni-

versitet, Trondheim,– Universitetet i Bergen,– Havforskningsinstituttet – Institute of Marine

Research (IMR), Bergen,– BCCR – Bjerknes Centre for Climate Research,

Bergen,– Havforskningsinstituttet, Austevoll, Storebø,– Sea Star International AS, Storebø,– Statoil ASA, Stavanger,– Miljostiftelsen Bellona, Oslo,– Miljoverndepartementet (Environment Min-

istry), Oslo,– Fiskeri- og Kystdepartementet (Ministry of Fish-

eries and Coastal Affairs), Oslo,– Olje- og Energidepartementet (Ministry of Petro-

leum and Energy), Oslo.Particular thanks are due to Statoil for allowing us tovisit its Sleipner gas production platform in the NorthSea and giving us a detailed introduction to the plat-form’s technology, gas production and CO2 seques-tration. Platform manager Egil Kai Elde was apatient guide, answering all our questions.

WBGU thanks Christopher Hay (TranslationBureau for Environmental Sciences, Seeheim-Jugen-heim, Germany) for his expert translation of thisreport into English from the German original.

VI

1

22.1

2.1.12.1.22.1.3

2.22.2.12.2.2

2.32.3.12.3.22.3.3

2.42.4.12.4.22.4.3

2.52.5.12.5.2

2.6

2.6.12.6.2

2.7

33.1

3.1.13.1.2

Acknowledgements V

Contents VII

Tables X

Figures XI

Boxes XII

Summary for policy-makers 1

Introduction 5

Global warming and marine ecosystems 7Climatic factors 7

Rising water temperatures 7Retreat of Arctic sea ice 8Changes in ocean currents 9

Impacts of global warming on marine ecosystems 11Natural climate variability 11Human-induced climate change 12

In focus: Climate and fisheries 16Changes in fish populations 16Regional prognoses of impacts on fisheries 17Global prognoses of impacts on fisheries 18

In focus: Climate and coral reefs 18Warming impact on corals 19Acidification impact on corals 20Measures for coral conservation 20

Guard rail: Conservation of marine ecosystems 21Recommended guard rail 21Rationale and feasibility 21

Recommendations for action: Improving the management of marine ecosystems 22

Fisheries management 23Marine protected areas 24

Research recommendations 30

Sea-level rise, hurricanes and coastal threats 33Climatic factors 33

Sea-level rise 33Stronger tropical cyclones 38

Contents

VIII

Impacts on coastal regions 40Biogeophysical impacts 40Socio-economic impacts 45

Guard rail: Sea-level rise 49Recommended guard rail 49Rationale 49Feasibility 51

Recommendations for action: Develop and implementadaptation strategies 52

Adapting coastal regions to the consequences of climate change 52The adoption of provisions governing loss of territory in international law 60


Ocean acidification 65Chemical changes in seawater 65

CO2 input 65Change in the carbonate budget 66Special role of CO2 67

Future development of the oceans as a carbon sink 67Effects of acidification on marine ecosystems 69

Physiological effects on marine organisms 69Effects on calcifying organisms 69Ecosystem structure and higher trophic levels 70Effects of acidification on fisheries 72Feedback of changes in calcification on the carbon cycle 72

Guard rail: Ocean acidification 72Proposed guard rail 72Rationale and feasibility 73

Recommendations for action: Linking climate protection with marine conservation 74

Reappraising the role of CO2 in climate protection policy 74Taking shipping sector emissions into account 75


CO2 storage in the ocean and under the sea floor 77CO2 sequestration 77

Potential and costs 77Risks and sustainability 78

Ocean storage 78Storage and residence time of CO2 79Impacts of CO2 storage on deep-sea organisms 79Present international law 80

Sub-seabed geological storage 80CO2 injection into the geological sub-seabed 80Risks and sustainability of CO2 storage in the seabed 81Regulating sub-seabed geological storage 82

Recommendations for action: Regulating CO2 storage 85Prohibiting CO2 injection into the ocean 85Limiting CO2 storage in the seabed 86


Methane hydrates in the sea floor 89The methane hydrate reservoir 89Methane release due to human intervention 90

Response to pressure and temperature changes 90

3.23.2.13.2.2

3.33.3.13.3.23.3.3

3.4

3.4.13.4.2

3.5

44.1

4.1.14.1.24.1.3

4.24.3

4.3.14.3.24.3.34.3.44.3.5

4.44.4.14.4.2

4.5

4.5.14.5.2

4.6

55.1

5.1.15.1.2

5.25.2.15.2.25.2.3

5.35.3.15.3.25.3.3

5.45.4.15.4.2

5.5

66.16.2

6.2.1

6.2.26.2.3

6.36.46.5

7

8

Effects of climate change on methane hydrates 90Mining of methane hydrates 92

Possible results of methane release 93Recommendations for action: Preventing methane release 94Research recommendations 95

Core messages 97

References 99

IX

Estimated global sea-level rise by the year 2300 with global warming limitedto 3°C 37Classification of damage caused by a rise in sea level 48Groups of calcifying marine organisms 70

Table 3.1-1

Table 3.2-1Table 4.3-1

Tables

Figure 2.1-1Figure 2.1-2

Figure 2.1-3Figure 2.1-4Figure 2.2-1Figure 2.2-2

Figure 2.3-1

Figure 2.6-1

Figure 3.1-1


Figure 3.2-1

Figure 3.2-2Figure 3.2-3Figure 3.2-4

Figure 3.2-5Figure 3.3-1Figure 4.1-1Figure 4.1-2Figure 4.1-3

Figure 4.3-1

Figure 4.4-1


Globally averaged sea-surface temperature, according to three data centres 8Development of sea-surface temperatures in the North Atlantic and European marginal seas 8Satellite photos of the Arctic ice cover 9The system of global ocean currents 10Schematic structure of a pelagic marine ecosystem 12Correlation of the catch of various economically important fish stocks withthe atmospheric circulation index 13Likely extension of the feeding area for some of the main fish populations if sea temperature increases 17Maritime zones under the United Nations Convention on the Law of the Sea (UNCLOS) 27Mean global temperature and sea level at different times in Earth’s history,with the projection for the year 2100 33Extent of melt area on Greenland according to satellite data 35The Larsen B ice shelf off the Antarctic Peninsula in satellite photographs 36Global sea-level rise as recorded by satellite measurements 37Temporal development of the energy of tropical storms and the average sea-surface temperature in the tropical Atlantic from August to October 39Distribution of land area, excluding Antarctica, as a function of elevation above present mean high water 40Coastal areas in Europe, parts of western Asia and North Africa 42Coastal areas along the North Sea 42Coastal areas along the Gulf of Bengal and in the Ganges-Brahmaputra-Meghna River Delta 43Population living below a certain elevation above mean high water in 1995 46Inundated areas in lower Manhattan (New York) 50Overview of the global carbon cycle 65Carbonate system of seawater 66Projections of different CO2 concentrations and their effects on the carbonate budget of the Southern Ocean 67Aragonite saturation and present occurrence of reef locations for warm-water corals 71Variability of the average pH value of the oceans in the past and present, as well as a projection for the future 73The Sleipner project in the North Sea 81Changes in the methane hydrate layer due to warming 90Causes and effects of methane hydrate destabilization 91Atmospheric methane concentration for a scenario with a total quantity of 1000Gt of anthropogenic CO2 emissions 93

Figures

The guard rail concept 6Maritime zones under international law 27Potential for conflict over resettlement 54Coastal management on the German North Sea coast 56The Sleipner project 81Marine renewables 83

Box 1-1Box 2.6-1Box 3.4-1Box 3.4-2Box 5.3-1Box 5.3-2

Boxes

The latest research findings show that climate changewill subject the marine environment and the coasts tomajor change and damage that are likely to havesevere consequences for humankind. Ocean surfacewaters are warming, the sea level is rising ever faster,the oceans are becoming increasingly acidic andmarine ecosystems are under threat. Human activi-ties are unleashing processes of change in the oceansthat are without precedent in the past several millionyears. Due to the considerable geophysical time lags,these processes will determine the state of theworld’s oceans for millennia to come. Humanity isthus intervening in a pivotal mechanism of the EarthSystem, and many of the consequences cannot yet bepredicted accurately. Resolute and forward-lookingaction is needed in order to ensure that the oceans donot overstep critical system boundaries. The way wehandle the oceans will be a decisive test of human-kind’s ability to steer a sustainable course in thefuture.

Climate mitigation for marine conservation

Ocean warming, ocean acidification and a distinctsea-level rise are all already measurable. The causesare clear: elevated concentrations of greenhousegases in the atmosphere caused by human activitieshave led to a global warming that has also increasedtemperatures in the surface waters of the oceans.Thisleads to rising sea levels due to thermal expansion ofthe water and due to melting ice masses. At the sametime, the continuously rising carbon dioxide concen-tration in the air causes CO2 to be absorbed by thesea where, through chemical reactions, the seawateracidifies. These changes can only be mitigated bymeans of drastic reductions in anthropogenic green-house gas emissions. Rapid action is thereforerequired:• Ambitious climate protection measures are

needed to limit the consequences of warming,acidification and sea-level rise for the marine envi-ronment and human society. WBGU thereforerecommends that global anthropogenic green-

house gas emissions must be approximatelyhalved by 2050 from 1990 levels. Adaptation mea-sures can only succeed if the present accelerationof sea-level rise and the increasing acidification ofthe oceans are halted.

• The guard rail already recommended previouslyby WBGU – namely limiting the rise in near-sur-face air temperature to a maximum of 2°C relativeto the pre-industrial value while also limiting therate of temperature change to a maximum of0.2°C per decade – is essential not only to preventdangerous climatic changes but also to maintainthe state of the oceans.

Bolstering the resilience of marine ecosystems

Compared to terrestrial ecosystems, marine ecosys-tems respond much more sensitively and rapidly toclimatic changes, for example through spatial shifts ofpopulations. As a result, human-induced warming ofthe surface waters can cause changes in food websand species composition that are difficult to predict.A further increase in water temperatures, in combi-nation with continuing acidification, will have majoroverall impacts on marine ecosystems and also onfisheries.

The fisheries sector is thus facing two furtherthreats in the future in the shape of climate changeand ocean acidification, in addition to the conse-quences of overfishing, which are already drasticenough in themselves. Taken together, and in view ofthe continuing growth of the world population, theseanthropogenic factors will jeopardize a sufficientsupply of food from the oceans.

Tropical coral reefs, by far the most species-richecosystems in the ocean, are acutely threatened byclimate change. Most reefs may be destroyed withinthe next 30–50 years, because many corals are notviable at higher water temperatures.The local ramifi-cations are vast, reefs being indispensable for coastalprotection and in supplying protein for millions ofpeople.

Summary for policy-makers


One of the most visible consequences of warmingis the retreat of Arctic sea ice. Over the past 30 years,summertime ice cover has declined by 15–20 percent. Model scenarios for the future indicate that,unless action is taken to mitigate climate change, theArctic Ocean will be practically ice-free in summer-time by the end of the 21st century. This would havesevere consequences for ecosystems and climaticprocesses.• To preserve marine biodiversity and strengthen

the resilience of marine ecosystems, WBGU pro-poses the following guard rail: at least 20–30 percent of the area of marine ecosystems should bedesignated for inclusion in an ecologically repre-sentative and effectively managed system of pro-tected areas. There is a particular need to enhancemarine conservation significantly for coral reefsand areas that are nursery grounds for fish popu-lations. Goals for marine protected areas alreadyagreed by the international community need to beimplemented, and the regulatory gap in thisregard for the high seas should be closed by adopt-ing an agreement under the United Nations Con-vention on the Law of the Sea (UNCLOS).

• Marine resource management should follow the‘ecosystem approach’. In particular, the publiclysubsidized overfishing of the oceans must be ter-minated, not least in order to strengthen theresilience of fish stocks to the impacts of climatechange. This necessitates not only removing fish-eries subsidies, but also dismantling excess fishingcapacity and taking measures to combat destruc-tive fishing practices and illegal or unregulatedfisheries.

• Our understanding of the linkages betweenanthropogenic disturbances, biological diversityand the resilience of marine ecosystems needs tobe improved. Intensive monitoring is a precondi-tion for the further development of coupledecosystem-climate models.

Limiting sea-level rise and reorienting coastalzone management strategies

Climate change causes sea-level to rise, particularlydue to ocean warming and the melting of inland glac-iers and continental ice sheets. Throughout the 20thcentury, global sea-level rise averaged 1.5–2.0cm perdecade. Satellite measurements show that thedecadal rate already reached 3cm in the past decade.If warming continues, there is a risk of further accel-eration of sea-level rise. There are indications thatthe continental ice sheets on Greenland and in theAntarctic are beginning to disintegrate. This has the

potential to cause several metres of sea-level rise inthe next centuries.

Besides sea-level rise, the increasingly destructiveforce of hurricanes is a further factor threateningmany coastal areas. Theory, observed data and math-ematical models agree that while climate warmingdoes not increase the number of hurricanes, it doesboost their destructive energy. Tropical sea-surfacetemperatures have warmed by only half a degreeCelsius, while an increase in the energy of hurricanesby 70 per cent has been observed.

Sea-level rise and extreme events such as hurri-canes and storm surges are threatening the coasts.Coastal protection is thus becoming a key challengefor society, not least in economic terms. Past strate-gies for protecting and utilizing coastal areas fail todo justice to this development. Novel combinationsof measures (portfolio strategies) are called for,whereby the options of protection, managed retreatand accommodation need to be weighed against eachother. In particular, coastal protection and natureconservation concerns must be better linked, and thepeople affected by adaptation or resettlement mea-sures need to be involved in decision-making on suchmeasures.• Guard rail: Absolute sea-level rise should not

exceed 1m in the long term, and the rate of riseshould remain below 5cm per decade at all times.Otherwise there is a high probability that humansociety and natural ecosystems will suffer unac-ceptable damage and loss.

• Because of anticipated sea-level rise, national andinternational strategies need to be developed forprotection and accommodation, but also for amanaged retreat from endangered areas.

• There is a need to improve the linking of natureconservation with coastal protection. The processof drawing up coastal protection plans and strate-gies for the sustainable use and development ofcoastal zones must integrate all key policy spheres(integrated coastal zone management).

Adopting innovative instruments of internationallaw for refugees from sea-level rise

Sea-level rise will lead to the inundation of coastsand small island states and thus to migration of ‘sea-level refugees’. Under international law as it standsat present, there is no obligation to receive refugeesfrom coastal areas, nor is the question about costsresolved. In the long term, however, the internationalcommunity will not be able to ignore the problem ofrefugees from coastal areas and will therefore needto develop appropriate instruments to ensure that

2

3Summary for policy-makers

affected people are received in suitable areas, ideallyareas corresponding to their preferences.• There is a need for agreements on the reception of

refugees from coastal areas and on the apportion-ment of the associated costs, e.g. by means of acompensation fund. It would be expedient todevelop a fair burden-sharing system, under whichstates make a binding commitment to assumeresponsibility for the migrants.

• To inform the policymaking process, studiesshould be undertaken in the fields of law andsocial sciences.

Halting ocean acidification in time

The dissolution of carbon dioxide in seawater leadsto considerable acidification (decrease in pH) andthus to changes to the biogeochemical carbonate bal-ance. The oceans have absorbed about one-third ofall anthropogenic CO2 emissions to date, which hasalready caused a significant acidification of seawater.Such emissions thus influence the marine environ-ment directly – in addition to the route via climatechange. Unabated continuation of this trend will leadto a level of ocean acidification that is without prece-dent in the past several million years and will be irre-versible for millennia. The effects upon marineecosystems cannot yet be forecast exactly but there isa risk of profound changes to the food web, as calci-fication of marine organisms may be impeded or insome cases even prevented. We are now seeing on aglobal scale problems similar to those that aroseregionally when lakes acidified in the 1970s and1980s (‘acid rain’).• In order to prevent disruption of the calcification

of marine organisms and the resultant risk of fun-damentally altering marine food webs, the follow-ing guard rail should be obeyed: the pH of nearsurface waters should not drop more than 0.2 unitsbelow the pre-industrial average value in anylarger ocean region (nor in the global mean).

• Engineering approaches to mitigate acidification,such as large-scale liming, are not feasible in theoceans. It is therefore important to ensure thatanthropogenic CO2 emissions are limited, regard-less of reductions of other greenhouse gas emis-sions.WBGU thus recommends taking the specialrole of CO2 compared to other greenhouse gasesinto account in the negotiations on future commit-ments under the United Nations Framework Con-vention on Climate Change. The consequences ofacidification for marine ecosystems and for bio-geochemical cycles are still insufficiently under-stood. Considerable further research is needed inthis regard.

Regulating CO2 storage

Engineering approaches can be used to capture thecarbon dioxide arising from the utilization of fossilenergy sources, and to compress it and transport it viapipelines or by ship to permanent repositories. CO2

can be stored in geological formations on land orunder the sea floor. Theoretically, the CO2 could alsobe injected into the deep sea. Such approaches, how-ever, involve a risk of continuous, slow release of thestored CO2 into the atmosphere, which runs counterto long-term climate mitigation. The specific benefitsand drawbacks of the technical and economic devel-opment of sequestration technologies therefore needto be balanced against other climate mitigationapproaches such as improving energy efficiency orswitching to renewable energy sources.• The precautionary principle indicates that intro-

ducing CO2 into seawater should be prohibited,because the risk of ecological damage cannot beassessed and the retention period in the oceans istoo short.

• Storing CO2 in geological formations under thesea floor can only be an ‘emergency’ solution for atransitional period. Permits for such measuresshould only be granted if they meet strict criteriawith regard to technical safety and, above all, withregard to the permanence of storage and its lowenvironmental impact. These criteria should alsoapply to the use of CO2 for ‘Enhanced Oil Recov-ery’. CO2 sequestration must not lead to neglect ofsustainable emissions reduction strategies (such asefficiency improvement and the promotion ofrenewable energies) and should therefore not besupported with public funds.

• Only a proportion of the CO2 stored under the seafloor should be eligible as prevented emissionswhen drawing up emissions inventories and forthe purposes of the flexible mechanisms in inter-national climate policy. This is necessary in orderto take the risk of leakage into account. Specificliability rules also need to be established.

Imposing strict conditions upon methane hydratemining

Quantities of carbon are stored in the sea floor in theform of methane hydrates that are of the order ofmagnitude of total worldwide coal reserves. Methanehydrates are only stable under high pressure and atlow temperature. Such conditions typically prevail onthe sea floor from depths of around 500m down-wards; in the Arctic this boundary is somewhathigher. The stability of methane hydrate stocks can


be compromised by climate change, by disturbancesresulting from mineral oil and natural gas produc-tion, or, in the future, possibly by direct mining of thehydrates themselves. WBGU takes the view that thehazard of a sudden release of larger, climate-relevantquantities of methane within this century is verysmall. Over the long term, however, the slow pene-tration of global warming to lower ocean layers andsediments could cause gradual methane releasesover many centuries to millennia.• Because of the potential instabilities of deposits, it

is important to ensure even now that methanehydrate mining in the oceans is only permittedunder very strict conditions. Existing regulatorysystems governing ocean mining should beamended and adjusted accordingly.

Complementing the existing financingmechanisms

Measures to mitigate and cope with the anticipatedadverse effects of climate change upon the marinehabitat can be funded from existing internationalfunds whose task is to finance emissions reductionsor adaptation projects. It must be expected, however,that these resources will not suffice for the tasks out-lined in the present report, above all because they donot budget for specifically ocean-related projects. Tocomplement these resources, WBGU therefore rec-ommends:• Fisheries subsidies must be removed in order to

avoid providing misplaced incentives for overfish-ing. The public funds thus released could then beinvested partly in marine conservation.

• Charges should be levied on the use of the oceansby shipping, and the revenues earmarked.

• The establishment of microinsurance systems toprotect individual assets should be supported as acomponent of a more comprehensive precaution-ary strategy, e.g. through public co-financing, espe-cially in developing countries.

• Some of the official development assistance(ODA) resources presently deployed to provideemergency relief worldwide should be divertedinto preventive measures.

With this special report, WBGU has taken up anissue that until now has attracted little attention, andwhose profound implications are largely underesti-mated.The state of the marine environment is of ele-mentary importance to the future of the blue planetEarth. Through overexploitation and pollution,humankind has already inflicted great damage on theoceans. Global climate change is presenting a further,

completely new dimension of threat. The presentreport pinpoints the threats and identifies requiredactions and options that arise at the interface of cli-mate change and the oceans. The report hopes toencourage policy-makers to tackle the necessarymeasures in time and with resolve, to prevent theoceans from becoming too warm, rising too high andturning too sour.

4

1Introduction

The oceans are changing rapidly. Surface waters arewarming, sea-level rise is accelerating and the oceansare becoming increasingly acidic, jeopardizing manymarine ecosystems. Human activities are unleashingprocesses of change in the oceans that are withoutprecedent in the past several million years. Humanityis thus interfering with pivotal mechanisms of theEarth System. The oceans play a key role in the car-bon cycle of our planet and have absorbed aboutone-third of total anthropogenic CO2 emissions untilnow. Covering more than two-thirds of the Earth’ssurface, the oceans initially take up the greater partof incoming solar heat and thus determine our cli-mate system. Similarly, the global water cycle is dri-ven mainly by evaporation from the oceans. Finally,the oceans harbour a great wealth of biological diver-sity and, through fisheries, supply humankind withvital proteins. An intact marine environment is alsoan important factor for economic development,social well-being and human quality of life.

Recent research is making it increasingly clearthat climate change will change and damage themarine environment and the coasts.These effects willalso impact severely upon human society.A large andgrowing part of the population now lives close tocoasts. The threats posed to coastal populations andinfrastructure by rising sea levels and extreme eventssuch as storm surges or hurricanes will mount in com-ing decades. Furthermore, together with drastic over-fishing, climate change and acidification can endan-ger food supply from the oceans. There is an urgentneed for action now in order to limit the adverseeffects of climate change upon ecosystems andhuman society, especially because, due to the consid-erable time lags, the present behaviour of humankindwill determine the state of the world’s oceans for mil-lennia to come. A strong research effort is alsoneeded, for the oceans are still terra incognita inmany respects.

One important reason to produce this specialreport is the changed scientific understanding of sea-level rise and ocean acidification since the Intergov-ernmental Panel on Climate Change published itslast assessment report (IPCC, 2001). Furthermore,

recent events such as the unusual hurricane season of2005, or the ongoing debate on methane hydratesand carbon storage, present a need for WBGU, theGerman Advisory Council on Global Change, tostate its views. By analysing the climatic impactsupon the oceans,WBGU draws attention to the needfor and urgency of efforts to engage in vigorous cli-mate mitigation activities and develop appropriateadaptation strategies. WBGU also wishes to con-tribute its findings to the process of shaping a newEuropean Union policy on seas and oceans.

This special report does not seek to paint a com-prehensive picture of the state of the oceans. It doesnot, for instance, set out to recapitulate the manyyears of debate on ocean overfishing. WBGU con-centrates instead on those key linkages between cli-mate change and the oceans that are the topic of newscientific insights.These insights include new findingson warming, ocean currents, sea-level rise, carbonuptake and acidification, and on the impacts of thesefactors upon marine ecosystems. The report also dis-cusses in detail the development of tropical cyclones,the issues surrounding carbon storage in the ocean orunder the seabed, and the risks associated withmethane hydrate deposits in the sea floor. Many ofthese issues are closely interlinked – coral reefs, forinstance, are affected simultaneously by warming,sea-level rise, storms and acidification. Each theme isexplored systematically, starting with the physicaland chemical fundamentals, proceeding to the eco-logical impacts, moving on to the consequences forhuman society, and finally deriving policy andresearch recommendations on that basis. WBGUembeds its analysis within a normative frameworkthat it has developed – the ‘guard rail’ approach (Box1-1). Analogous to the ‘climate guard rail’ that itdeveloped previously, WBGU now proposes a set of‘ocean guard rails’ for the sustainable managementof the oceans.These are quantitative boundaries thatmust not be overstepped.

Resolute and forward-looking action is needed toensure that the oceans do not cross critical systemboundaries within a matter of decades. Oversteppingthese boundaries would lead to severe and partly

irreversible damage to nature and human society.The way we manage the oceans now will thus be adecisive test of humankind’s ability to steer a sus-tainable course in the future.

6

BBooxx 11--11

TThhee gguuaarrdd rraaiill ccoonncceepptt

WBGU has developed the idea of guard rails to opera-tionalize the concept of sustainable development (e.g.WBGU, 2004). Guard rails are limits on damage and can bedefined quantitatively; a breach of these limits would giverise either immediately or in future to intolerable conse-quences so significant that even major utility gains in otherfields could not compensate for the damage. Guard railsthus demarcate the realm of desirable and sustainabledevelopment trajectories. For instance, WBGU has arguedrepeatedly in previous reports that the average mean tem-perature should not be allowed to rise more than 2°C abovethe pre-industrial level. Beyond that value, a domain of cli-mate change begins that is characterized by non-tolerabledevelopments and risks.

The guard rail approach proceeds from the realizationthat it is scarcely possible to define a desirable and sustain-able future in positive terms, in other words as a specific tar-get or state that should be achieved. It is, however, possibleto agree on the demarcation of a domain that is recognizedas unacceptable and which society wishes to prevent.Withinthe guard rails, there are no further requirements at first.Society can develop in the free interplay of forces. Only if asystem is on course for collision with a guard rail must mea-sures be taken to prevent it crossing the rail. Compliancewith all guard rails does not mean, however, that all socio-economic abuses and ecological damage will be prevented,as global guard rails cannot take account of all regional andsectoral impacts of global change. Moreover, knowledge islimited and misjudgement is possible. Compliance withguard rails is therefore a necessary criterion for sustainabil-ity, but it is not a sufficient one.

The analogy of road traffic may serve to illustrate theguard rail concept. Guard rails have a function similar tothat of speed limits, e.g. a limit permitting a maximum of50km per hour in built-up areas.The outcome of setting thelimit at 40, 50 or 60km per hour can be determined empiri-cally, but in the final analysis the choice of figure is a nor-mative decision, representing an expedient way to handle arisk collectively. Compliance with the speed limit cannotguarantee that no serious accidents will occur, but it cankeep the risk within boundaries accepted by society. The

guard rails formulated by WBGU build upon fundamentalnorms and principles agreed by the international commu-nity in various forms. They can be no more than proposals,however, for the task of defining non-tolerable impacts can-not be left to science alone. Instead, it should be performed– with the support of scientists – as part of a worldwide,democratic decision-making process. For instance, compli-ance with the climate guard rail (no more than 2°C globalwarming) has now been adopted as a goal by the EuropeanUnion.

Guard rails for marine conservationIn the present report, WBGU applies its guard railapproach to the field of marine conservation. This buildsupon earlier reports, in which WBGU has repeatedlyargued for a two-fold climate guard rail (WBGU, 1995,2003). The environmental changes in the oceans discussedin this report further underpin the need for the climateguard rail. In addition, the report develops further guardrails. Each is concerned with a specific aspect of the inter-play between climate change and the oceans, and is eluci-dated and argued in a separate chapter.The full set of guardrails is as follows:• Climate protection: The mean global rise in near-surface

air temperature must be limited to a maximum of 2°Crelative to the pre-industrial value while also limiting therate of temperature change to a maximum of 0.2°C perdecade.The impacts of climatic changes that would ariseif these limits are exceeded would also be intolerable forreasons of marine conservation.

• Marine ecosystems: At least 20–30 per cent of the area ofmarine ecosystems should be designated for inclusion inan ecologically representative and effectively managedsystem of protected areas.

• Sea-level rise: Absolute sea-level rise should not exceed1m in the long term, and the rate of rise should remainbelow 5cm per decade at all times. Otherwise there is ahigh probability that human society and natural ecosys-tems would suffer non-tolerable damage and loss.

• Ocean acidification: In order to prevent disruption ofcalcification of marine organisms and the resultant riskof fundamentally altering marine food webs, the follow-ing guard rail should be obeyed: the pH of near surfacewaters should not drop more than 0.2 units below thepre-industrial average value in any larger ocean region(nor in the global mean).

Introduction

2.1Climatic factors

2.1.1Rising water temperatures

The temperatures in the ocean influence sealife aswell as the solubility of carbon dioxide in the water.They change the density of seawater, thereby influ-encing the currents and the sea level: the thermalexpansion of water contributes considerably to sea-level rise. The sea surface temperature also affectsthe atmosphere in a multitude of ways. The mildAtlantic air that is often felt in Europe during thewinter obtains its heat from the relatively warmocean. High water temperatures also lead toincreased evaporation, which is an important energysource for the atmosphere (for example, in tropicalcyclones) and a source of water for many intensiveprecipitation events (among others, the Elbe riverflood of 2002 in Germany).

Significantly improved data sets of global oceantemperatures covering the past 50 years havebecome available to researchers in recent yearsthrough international efforts in the exchange of data(NODC, 2001). Based on over 7 million measuredtemperature profiles, Levitus et al. (2005) havereconstructed a time series of the heat content of theworld ocean. They report an increase in the amountof stored heat of 15.1022 joules from 1955 to 1998.This corresponds to an average heat absorption of 0.2watts per m2 for this time period when averagedacross the entire surface of the Earth. For the periodfrom 1993 to 2003 heat absorption was even greater,at 0.6 watts per m2 (Willis et al., 2004). This increaseof heat in the ocean indicates that the Earth ispresently absorbing more energy from the sun than itcan radiate back into space.This reveals a state of dis-equilibrium in the heat budget of the Earth, as is tobe expected due to the anthropogenic greenhouseeffect (Hansen et al., 2005).

Averaged globally and throughout the entirewater column, the temperature of the ocean has onlyrisen by 0.04ºC since 1955. So far only the surfacemixed layer with a thickness of a few hundred metreshas warmed, while the average ocean depth is 3800m.The amount of sea-level rise caused by thermalexpansion of the water so far is therefore only a smallfraction of what will result when the warmingextends into the deep sea over the coming centuries(Section 3.1.1).

Figure 2.1-1 shows the variation of the sea-surfacetemperature, which is very important for the climatesystem. It shows a strong similarity to the develop-ment of air temperatures, but the warming is not aspronounced (0.6°C since the beginning of the twenti-eth century).These two facts are not surprising.Ther-mally, the sea surface is closely coupled to the over-lying atmosphere. Making up 30 per cent of theEarth’s surface, the land masses, because of theirlower heat capacity, warm up more quickly than theoceans, so the global mean air temperature rises gen-erally more quickly than that of the ocean.A data setof air temperatures measured by ships at night abovethe sea surface (Parker et al., 1995) shows a patternvery similar to the water temperatures. These datasupport the fact of a warming trend in the ocean sur-face waters and once more confirm the global warm-ing measured by weather stations.

Figure 2.1-2 shows the increase in surface temper-atures in the North Atlantic, which in large part rangebetween 0.3 and 1°C over the indicated time period.A significantly stronger warming of several degreesis seen in Arctic latitudes, primarily because of thepositive (strengthening) feedback with the shrinkingsea ice (Section 2.1.1). Some small areas, however,show a cooling trend due to dynamic changes in thesea.This is particularly true of the Gulf Stream regionoff the coast of the USA and in regions near Green-land. The reason is probably natural internal fluctua-tions in the circulation, which superimpose the gen-eral warming trend caused by greenhouse gases.

The increase of sea temperatures in tropical lati-tudes is of particular interest because it influencestropical storms.This will be discussed in Section 3.1.2.

Global warming and marine ecosystems 2

2 Global warming and marine ecosystems

2.1.2Retreat of Arctic sea ice

An especially strong warming of seawater has beenobserved in the Arctic region in recent decades. Thiswas described in 2004 in detail along with its impactsin an international study (Arctic Climate ImpactAssessment; ACIA, 2005).

The study concludes that a significant reduction ofthe Arctic sea ice has occurred that can not beexplained by natural processes but only by humaninfluences.The ice retreat can be clearly seen in satel-lite photographs (Fig. 2.1-3). The satellite time seriesfrom 1979 to 2005 shows a decline in the ice area of15 to 20 per cent. The lowest ice extent ever mea-sured was recorded in September 2005. Using a com-pilation of observations from ships and coastal sta-tions, this development can be extended back to the

time before satellite measurements were available.These kinds of observations go back to the year 1900,and cover about 77 per cent of the area of the Arcticregion. The long-term data strongly suggest that thepresent shrinking of the ice cover is a unique event inthe past hundred years.

Changes in the thickness of Arctic ice are moredifficult to observe than its lateral extent. With theend of the Cold War, measurements by military sub-marines that had patrolled beneath the Arctic icebecame available. These data indicate that the icethickness may have already decreased by 40 per cent(Rothrock et al., 1999). Other investigations suggestsmaller decreases in the thickness. Johannessen et al.(2005) report a decrease of 8–15 per cent, so theactual changes still have to be regarded as uncertain.

Additional knowledge for the Arctic Ocean isobtained from computer models with high spatialresolution, driven by observed weather data. For

8

FFiigguurree 22..11--22Development of sea-surfacetemperatures in the NorthAtlantic and Europeanmarginal seas. Temperaturechanges of the yearlyaverage between 1978 and2002 are shown (as a lineartrend). Based on the GISSTdata set of the BritishHadley Centre.Source: PIK, based onHadley Centre, 2003

0.4

0.2

0

-0.2

-0.4

-0.618801860 19201900 1940 1960 1980 2000

Year

Dev

iatio

n fr

om m

ean

tem

per

atur

e 19

61–1

990

[°C

]UKMO

NCDC

NCEP

FFiigguurree 22..11--11Globally averaged sea-surface temperature,according to three datacentres: The UK Met Office(UKMO, blue), the USNational Center forEnvironmental Prediction(NCEP, black), and the USNational Climatic DataCenter (NCDC, red).Source: IPCC, 2001a

9Climatic factors 2.1

recent decades they show a decrease in ice extentthat is in agreement with the satellite data as alreadydiscussed. In these models the ice thickness decreasesmore strongly, about 43 per cent since 1988 (Lindsayand Zhang, 2005). Maslowski et al. (2005) obtainedsimilar numbers. If the warming continues un-checked, the scenarios produced by global modelsindicate that the Arctic Ocean will be practically ice-free in the summer by the end of this century (MPIfür Meteorologie, 2005). According to the regionalmodels mentioned, this condition could occur evenearlier.

2.1.3Changes in ocean currents

Since the 1980s scientists have begun to address thequestion of possible abrupt changes in Atlantic cur-rents and their effects on the climate (Broecker,1987). The basic problem – a possibly strong nonlin-ear response of the current to freshwater influx – was

recognized as early as the 1960s (Stommel, 1961). Inrecent years there has been an increased focus byresearchers on the probability and the possibleimpacts of such events. However, the research is stillat an early stage and many questions have not yetbeen answered. The danger of changes in the marinecurrents was brought to the attention of the publicthrough the ‘Pentagon Report’ by Schwarz and Ran-dall (2003), which featured in the media in 2004.Thisreport presented a worst-case scenario in which, dur-ing the next 10 to 20 years, the North Atlantic Cur-rent stops flowing, which would lead to a severe cool-ing in the North Atlantic region within just a fewyears. This is, however, a speculative and extremelyimprobable scenario. In the present situation there isno evidence to support an imminent change in thecurrents. But in the longer term, and with continuedclimate warming, this could develop into a seriousdanger by the middle of this century.

Huge masses of water currently sink from the sur-face to great depths in the Nordic Seas and theLabrador Sea. From there the water flows south-wards at depths of 2–3km to the Southern Ocean(Figure 2.1-4). Balancing this loss of water, warm sur-face water flows from the south into the northern lat-itude regions.This results in a large-scale turnover ofwater in the Atlantic, in which around 15 million m3

of water per second are transported. Like a central-heating unit, the ocean transports 1015 watts of heatto the northern Atlantic region through this process,which is equivalent to 2000 times the total output ofEurope’s power stations.

Global climate change affects this water flow bydecreasing the density of seawater in two ways: first,the temperature increase causes a thermal expansionof the water and, secondly, increased precipitationand meltwater input dilute the seawater with fresh-water.This density decrease can retard the sinking ofwater in the northern Atlantic, the so-called deep-water formation. Particularly in the Nordic Seas asalinity decrease has already been observed in recentdecades (Curry and Mauritzen, 2005), althoughaccording to modelling calculations this trend is stilltoo weak to have an impact on Atlantic current pat-terns.

British researchers have recently reported mea-surements suggesting that the circulation in theAtlantic may have already weakened by 30 per cent(Bryden et al., 2005). The interpretation of thesedata, however, is still contested in professional cir-cles, in part because they do not agree with modellingcalculations or with changes in sea-surface tempera-tures (Figure 2.1-2), where such a weakening of heattransport should be accompanied by a noticeablecooling. But if the trends of warming and salinitydecrease should continue to strengthen in the coming

FFiigguurree 22..11--33Satellite photos of the Arctic ice cover, (a) September 1979and (b) September 2005.Source: NASA, 2005


decades, this may actually lead to a noticeable weak-ening of the Atlantic current over the course of thiscentury, and in an extreme case possibly even to atotal cessation of deep-water formation.

In all probability the consequences would besevere. The North Atlantic Current (not the GulfStream as is often too simply stated) and the greaterpart of the Atlantic heat transport would be shutdown. This would significantly change the tempera-ture distribution throughout the entire Atlanticregion. Depending on the degree of warming that hastaken place before, it could even lead to regionalcooling to levels below today’s temperatures. South-ern Hemispheric warming would then be evenstronger.

As a result of dynamic adaptation of the sea sur-face to the altered currents, sea level in the NorthAtlantic would quickly rise by up to 1m and slightlyfall in the Southern Hemisphere. This redistributionof water would not have an immediate impact on theglobal sea-level average (Levermann et al., 2005).But over the long term the global average would riseby an additional 0.5m due to the gradual warming ofthe deep ocean after the loss of input of cold water.In addition, the tropical precipitation belt would verylikely shift because the ‘thermal equator’ would driftsouthward (Claussen et al., 2003). This is indicatedboth by model simulations and historical climatedata.

Initial simulation computations also show a reduc-tion of the plankton biomass in the Atlantic by half(Schmittner, 2005; Section 2.2.2.2). Because of ther-mohaline circulation the Atlantic is presently one ofthe most fertile marine regions and most productivefisheries areas of the Earth. In addition, the interrup-tion of deep-water formation would reduce theocean’s uptake of anthropogenic CO2 (Chapter 4).

A breakdown of the North Atlantic Current is arisk that is difficult to calculate, but which would havesevere adverse effects. One critical factor is theamount of freshwater that enters the northernAtlantic in the future. This will depend in large parton the speed at which Greenland’s ice sheet melts. Areliable prediction is not possible with the presentstate of knowledge; at best, a risk estimation can beattempted. For this purpose the Potsdam Institute forClimate Impact Research together with the Ameri-can Carnegie Mellon University questioned a dozenof the world’s leading experts in the autumn of 2004,in detailed interviews lasting around six hours each.Their estimations of the risk of a total stop of deep-water formation and the associated currents variedconsiderably, but some were surprisingly high (Zick-feld et al., submitted).With an assumed global warm-ing of only 2ºC by the year 2100, four of the expertsestimated the risk at greater than 5 per cent; with3–5ºC of warming, four of the experts indicated a riskexceeding 50 per cent.

10

Surface flow Bottom flow

Deep flow

Salinity > 36 ‰

Salinity < 34 ‰Deep water formation

FFiigguurree 22..11--44The system of global ocean currents, primarily showing the ‘thermohaline’ circulation that is driven by temperature and salinitydifferences.Source: after Rahmstorf, 2002

11Impacts of global warming on marine ecosystems 2.2

2.2Impacts of global warming on marine ecosystems

This section focuses on the impacts of climate warm-ing (see Section 2.1) on marine ecosystems. WBGUconsiders this to include the entire marine realm,from the high seas to aquatically dominated coastalecosystems. WBGU has deliberately only selectedfactors that are important to the subject of this spe-cial report. Overfishing, considered to be the mostsignificant adverse anthropogenic impact today(Pauly et al., 2002; MA, 2005b), is not discussed.Alsonot treated here are direct destruction of marineecosystems, pollution and alien species invasions(GESAMP, 2001; UNEP, 2002). Acidification of thesea is treated in Chapter 4. Togehter, these anthro-pogenic impacts have already strongly reduced theresilience of many marine ecosystems (Jackson et al.,2001).

The most productive areas in the oceans, the shal-low continental shelves (<200m water depth) are themost intensely affected by these impacts. Althoughthe shelves make up less than 7 per cent of theocean’s surface, this is where the greatest proportionof the primary and secondary production takes place,and where the most productive fishing grounds arefound (Section 2.3). The primary production of theseas by algae (phytoplankton) is limited to thetranslucent upper water layer, the euphotic zone(down to approx. 200m water depth). A multitude ofsecondary producers live from these primary produc-ers, especially zooplankton, fish and marine mam-mals, both in open water (pelagic) and at or belowthe sea floor (benthic). All organisms are linked toone another through a complex food web (Figure 2.2-1). For its energy source, the fauna of the dark deepsea is dependent on the organic carbon from the pri-mary production, which sinks to the depths as deadbiomass (‘biological pump’). Only in the vicinity ofhydrothermal vents in the deep sea do bacteria forman independent basis for higher life forms throughchemosynthesis.

The coastal ecosystems are also of great biologicaland economical importance. In addition to their eco-nomic utility, some species-rich coastal systems suchas wetlands, mangrove forests and coral reefs play aspecial role in protecting the coasts from floodingand erosion (Section 3.2).

2.2.1Natural climate variability

The natural variability of abiotic factors in marineecosystems such as water temperature or ocean cur-

rents is relatively great, and often follows non-linearor cyclic patterns. Studying the effects of natural cli-mate variability can provide valuable informationabout the impacts of global warming. Compared toterrestrial systems, marine ecosystems react moresensitively and quickly to changes in climatic condi-tions, with unpredictable consequences for speciescompositions, spatial shifts of populations, or restruc-tured food webs (Steele, 1998; Hsieh et al., 2005;overview by Brander, 2005). As Klyashtorin (2001)has shown, many Atlantic and Pacific fish stocksexhibit close correlations with climate patterns overmany decades (Figure 2.2-2), for example, with theatmospheric circulation index, which describesatmospheric conditions in the Atlantic-Eurasianregion. Even small natural climatic changes can havesignificant effects on marine ecosystems and fishstocks – through direct temperature effects, as aresult of changes in primary production, or throughimpacts on important development stages (e.g., juve-nile fish stages:Attrill and Power, 2002). For example,the cod stocks off Greenland reacted to a warming ofthe North Atlantic in the 1920s and 1930s with a rapidexpansion to the north (approx. 50km per year) anda considerable increase of stock size, which laterdecreased again as a result of overfishing and deteri-orating climatic conditions (Jensen, 1939). Plankton-feeding fish species in particular, such as sardines oranchovies, show strong natural stock fluctuations, inwhich large-scale climatic variations play an impor-tant role (Barber, 2001; PICES, 2004).The short-termdisturbances of the ENSO events (El Niño/SouthernOscillation), for example, have far-reaching, 2- to 3-year effects on the marine ecosystems of the Peru-Humboldt current system (decreased nutrient supplycausing lower primary production, partial collapse offish populations: Barber, 2001) and on the most pro-ductive fish stock in the world (Peruvian anchovies:FAO, 2004; Bertrand et al., 2004). The impacts of theENSO events are, however, reversible, with ‘normal’conditions being re-established as a rule within a fewyears (Fiedler, 2002).

Ignoring small interannual variations, however,regional climatic conditions, along with the structureand dynamics of the ecosystems within a marineregion, can also remain relatively stable over a periodof many years or decades, defining what is generallyreferred to as a regime. When this kind of relativelystable situation changes rapidly, within the course ofone or two years, it is called a ‘regime shift’ (King,2005). Along with these regime shifts, considerablestructural changes in the affected marine ecosystemoccur, from the phytoplankton up to the highesttrophic levels in the food web, including large preda-tory fish.

12

Regime shifts have been observed often and invarious marine regions (King, 2005). In the North Seain the late 1980s, for example, a regime shift occurredthat was related to abrupt changes in surface temper-ature, wind conditions and a multitude of biologicalparameters (Reid et al., 2001; Beaugrand, 2004;Alheit et al., 2005). Due to an increase in westerlywinds the influx of warm water into the North Seawas strengthened causing, among other things, adegradation of living conditions for North Sea cod.There is probably a connection between this persis-tent change in the North Atlantic Oscillation andanthropogenic climate warming (Gillett et al., 2003).In the North Pacific off the coast of California, alter-nating regimes with a period of around 60 years havebeen documented covering almost two millennia(Baumgartner et al., 1992). They cause a distinctrestructuring of the marine ecosystems (Hare andMantua, 2000; King, 2005).

How regime shifts are triggered and what effectsthey have in the food web of an ecosystem are not yetthoroughly understood, even though quite detailedobservations of changing ecosystem structures doexist. The energy fluxes originating in the phyto-plankton, at the base of the food web, often seem toplay an important role (‘bottom up’ control: e.g.,Richardson and Schoeman, 2004). However, struc-tural changes can also be controlled ‘top down’,caused by the collapse of the population of predatoryfish, either by overfishing (Worm and Myers, 2003;Frank et al., 2005) or by climatic changes (Polovina,

2005), and reaching down to the lower levels of thefood web by trophic coupling.

2.2.2Human-induced climate change

Although the natural variability can be very largeregionally, the global warming trend already pre-dominates in most areas (Figure 2.1-2). The anthro-pogenic impact on various climatic factors hasalready had observable effects on the distribution ofmarine organisms and the species assemblages ofmarine ecosystems (overview by Brander, 2005). Cli-mate impacts have been described for all levels of theecosystem, from primary production (Section 2.2.2.2)to zooplankton (e.g., Richardson and Schoeman,2004) and small pelagic fish species (sardines), all theway up to the large predatory fish (tropical tuna:Lehodey et al., 2003).

2.2.2.1Effects of water temperature on the physiology ofmarine organisms

According to the latest findings, temperature has asignificantly greater influence on the distribution ofanimal and plant species than was previouslyassumed, and this is independent of the position ofthe organisms in the food web (Huntley et al., 2004).

Nutrients

Bacteria

Particulate organic matter

Carbonate system

CO2

Sediment

Light

FishZooplanktonPhyto-plankton

Predatoryfish

FFiigguurree 22..22--11Schematic structure of apelagic marine ecosystem.Green arrows: input toprimary production; blackarrows: interaction with thecarbonate system; brownarrows: decomposition ofbiomass. In the interest ofclarity, marine mammals andseabirds are not shown.Source: WBGU



To a large degree, the window of thermal tolerance inwhich a species can survive, grow and reproducedetermines its distribution (Pörtner, 2005). Anincrease in water temperature (Section 2.1.1) influ-ences the life of marine organisms both directly andindirectly.A direct physiological impact is seen whenthe upper limit of the temperature tolerance rangefor a species is exceeded.This applies, for example, totropical corals (Section 2.4.1). An indirect influenceof increasing water temperature is observed, forexample, when organisms previously available tem-porally and spatially as food for a species are nolonger present due to changes in the species assem-blage of an ecosystem caused by temperature differ-ences (Section 2.2.2.5). Both of these effect chainscan lead to shifts of populations, the invasion of alienspecies, and even the disappearance of species.

2.2.2.2Phytoplankton and global primary production

The climatic factors altered by human activities (Sec-tion 2.1) initially affect the phytoplankton and there-fore primary production. The total marine ecosys-tem, all the way up through the various trophic levelsto the large predators such as tuna and sharks, feedsin principle from the primary production. Therefore,

through this coupling, a change in primary produc-tion will have an effect on the higher trophic levels ofthe food web and will be reflected in changed speciesassemblages or biomasses in the total ecosystem.Theprimary production is influenced by many climaticfactors (Fasham, 2003):• Temperature: Growth and species composition of

the phytoplankton are strongly dependent ontemperature. Initially, primary production isdirectly stimulated through warming. But theincreased temperature can also indirectly slowdown production, for example due to a decrease innutrient supply resulting from prominent temper-ature stratification.

• Light: Changes in the ice or cloud cover of the sur-face water have a direct influence on the primaryproduction because the phytoplankton requirelight as an energy source.The light supply for phy-toplankton also diminishes with increased mixingof the surface water.

• Nutrients: Climate change can also indirectlyinfluence the supply of nutrients to the phyto-plankton (primarily nitrogen and phosphorous,but also ‘micronutrients’ such as iron: Jickells etal., 2005). Through the sinking of dead organismsfrom the productive upper layer of the ocean,organic material and thereby nutrients are contin-uously exported to the deep sea (‘biological

Pacific salmonCalifornian sardineJapanese sardineAlaska pollockEuropean sardinePeruvian sardine

AtmosphericCirculation Index

1.0

0.8

0.6

0.4

0.2

0

200

100

-500

-400

-300

-200

-100

0

19201900 1940 1960 1980 2000

Year

Cat

ch, r

elat

ive

units

Atm

osp

heric

Circ

ulat

ion

Ind

ex

FFiigguurree 22..22--22Correlation of the catch of various economically important fish stocks with the atmospheric circulation index.Source: compiled on the basis of Klyashtorin, 2001


pump’: Falkowski et al., 2003). The return trans-port to the upper layers occurs through upwellingcurrents and vertical mixing, which are influencedby climate in the form of temperature stratifica-tion, as well as wind and current conditions (forexample, Sarmiento et al., 2003).

In addition to all these factors, climate warming islargely a result of rising CO2 concentrations. In manyphytoplankton species this leads to a direct increasein the rate of photosynthesis, although the variousspecies groups benefit to differing degrees (Section4.3.1). These various factors are also all coupled withone another. The warming of the surface layers notonly increases photosynthesis rates, it also promotesa more stable layering of the water column, whichdecreases the nutrient supply and weakens theplankton production. The stronger stratification canalso destabilize the dynamics of phytoplankton pro-duction (Huisman et al., 2006). An increased windspeed, on the other hand, counteracts the tempera-ture effect upon stratification. In the northeastAtlantic the sum of these counteracting effects pro-duces a net increase of phytoplankton in cold-waterregions (because here with the good nutrient supplyand higher turbulence the improved metabolismrates due to temperature increase are predominant),and a net decrease in warm-water regions (becausestronger stratification, under limited nutrient avail-ability, worsens the growing conditions; Richardsonand Schoeman, 2004). It is therefore not surprisingthat these effects are difficult to model and varygreatly from region to region.

Satellite-based observations of the phytoplanktonbiomasses derived from the chlorophyll content ofseawater reveal that the global annual primary pro-duction has decreased in nine of twelve oceanregions since the 1980s, and the global average bymore than 6 per cent (Gregg et al., 2003). The highnorthern latitudes account for 70 per cent of theglobal decline, presumably caused by the worseningnutrient supply due to the rise in temperature. Onlythree tropical ocean regions (northern and equator-ial Indian and the equatorial Atlantic) exhibited anincrease. For the North Atlantic, long-term dataseries based on physical samples show an increase inphytoplankton north of 55°N and a decrease south of50°N (Richardson and Schoeman, 2004). The projec-tions for a future with global warming show contra-dictory trends. The modelling of Bopp et al. (2001)suggests a reduction of the global marine export pro-duction (which correlates well with primary produc-tion) by about 6 per cent in the next 65–75 years witha doubling of the atmospheric CO2 concentrations.Production in the tropics would decline due tostronger stratification and the resulting decrease innutrient supply, while increasing in the subpolar

regions. In contrast, the models of Sarmiento et al.(2004), with large uncertainty, show a slight increasein the global primary production. Again, the effectsare regionally highly variable. The authors of theArctic Climate Impact Assessment consider it prob-able that moderate warming would promote primaryproduction in the Arctic, mainly due to the reductionof sea ice (ACIA, 2005).

So the available findings are, at least in part, con-tradictory, and regional observations are not alwaysin agreement with model prognoses. Obviously, ourunderstanding of the critical processes, such as thetemperature sensitivity of primary production, isinsufficient. The quality of coupled climate, ocean,and ecosystem models presently does not allow anyrobust conclusions (Sarmiento et al., 2004), althoughsome regional models are already able to representthe connections between the changes in ocean cur-rents and primary production (examples in Brander,2005).

It is improbable that climate change will lead tothe breakdown of the North Atlantic Current, butthis possibility cannot be excluded (Section 2.1.3;Rahmstorf, 2000; Curry and Mauritzen, 2005). Thesimulations of Schmittner (2005) show a completelyaltered ecosystem situation for this scenario: the bio-masses of phyto- and zooplankton in the NorthAtlantic would decrease by half due to sharplyreduced nutrient supply in the surface waters, withcorresponding large impacts on ecosystem produc-tivity and structure.

2.2.2.3Zooplankton

Primary production by phytoplankton is the nutri-tional basis for the zooplankton (secondary produc-tion: often small crustaceans), which is in turn signif-icant as food for the growth of fish populations. Fishlarvae in particular are dependent on the synchro-nous and high availability of appropriate zooplank-ton, so that fish stocks can replenish and productionis maintained. The following examples show that forthe zooplankton too, noticeable changes can alreadybe identified as a result of anthropogenic climatechange.

In the North Atlantic the distribution of copepods,an important group in the marine food web, hasshifted to the north by around 10° of latitude as aresult of a combination of changes in the NorthAtlantic Oscillation (NAO) and human-induced cli-mate change (Beaugrand et al., 2002). For the NorthSea cod these changes, along with overfishing, havecontributed to poor conditions encountered by the

14


fish larvae and a steady decline of the population(Beaugrand et al., 2003).

Krill (Euphausia superba) in the Antarctic havedeclined significantly since 1976, while other zoo-plankton species (salps) have increased, which canprobably be attributed to the climate-driven reduc-tion of sea ice around the Antarctic peninsula(Atkinson et al., 2004). Because krill is an importantfood source for fish, penguins, seals and whales, thishas led to significant changes in the food web in theSouthern Ocean. Investigations of planktonicforaminifera in sediments covering the past 1400years have revealed an anomalous change in thespecies assemblages in recent decades. This suggeststhat the anthropogenic warming of the ocean hasalready exceeded the range of natural variability(Field et al., 2006).

2.2.2.4Marine mammals

The warming also causes a decrease in the geo-graphic extent of the Arctic sea ice. This especiallyaffects animals such as polar bears and ringed seals,which are directly dependent on this habitat in theirfeeding habits and for the rearing of their young(ACIA, 2005).

Polar bears feed almost exclusively on seals, whichare bound to the ice habitat. Female polar bears beartheir young in caves on the land. In the spring aftertheir winter sleep, in order to reach their huntingareas on the ice, the mother and her young aredependent on ice corridors, because the young ani-mals cannot cross large areas of open water. If the icecontinues to recede, they will not be able to reachtheir hunting grounds. Adult polar bears are goodswimmers, and can cover distances in the water ofover 100km. Monnett et al. (2005), however, report adoubling of the number of polar bears sighted swim-ming in open water within a 20-year observationperiod, as well as most recent finds of four drownedpolar bears near Alaska in a location where the icewas over 200km to the north of its normal seasonallimit.Around Canada’s Hudson Bay, the area of theirsouthernmost occurrence, the polar bear populationhas declined by 22 per cent since 1987 (Carlton,2005). With the loss of the summer sea-ice cover,polar bears are forced into a life on land, where theyencounter competition with brown and grizzly bearsand increased contact with humans, which reducesthe chances for survival of this species.

Scenarios for the Baltic Sea also indicate that theice cover here will significantly decrease over thenext 30 years. The Baltic ringed seal requires a firmice layer with a snow cover for at least two months for

rearing its young. Of the four former breeding areasin the Baltic Sea with separate populations only onesuitable area will remain available in the future: thenorthern Bay of Bothnia (Meier et al., 2004). Initialobservations have been reported in the Antarctic,too, that can be attributed to climatic changes. For thepast 20 years birth rates of the cape fur seal havebeen in decline. This decrease correlates with unusu-ally high temperatures of the surface water subse-quent to the abundant El Niño events between 1987and 1998, and it has presumably been intensified bythe lowered nutrient supply – primarily krill – for thefemale seals (Forcada et al., 2005).

These examples illustrate how changes in the icehabitat caused by climatic change can drasticallyimpact the highest trophic levels.

2.2.2.5Ecosystem impacts

Temperature increases and other factors related toclimate change affect groups of organisms in differ-ent ways, so that population shifts can occur at differ-ent rates and intensities to separate species that pre-viously inhabited the same region or were present atthe same time. This decoupling of previously syn-chronous trophic levels (‘trophic mismatch’) can pro-duce considerable changes in the ecosystem struc-ture (for example, in the North Sea: Edwards andRichardson, 2004). Climate-induced spatial changesin the phytoplankton distribution can affect both theherbivorous zooplankton and the carnivorous zoo-plankton, so that fish, seabirds and mammals alsohave to adapt to the new conditions (Richardson andSchoeman, 2004).

These kinds of large-scale shifts have already beenobserved at different levels in the food web, forexample in the North Atlantic (Beaugrand and Reid,2003). After an anomalous temperature increase inthe 1980s, populations of cold-water species such aseuphausids and copepods shifted northward and thestocks decreased, while the smaller warm-waterspecies showed a corresponding increase (Beau-grand et al., 2002). This then led to a decline in thesalmon population. For the future, Beaugrand andReid (2003) expect a continuing decline in the num-ber and distribution of the salmon population, espe-cially at the southern edge of its geographic distribu-tion (Spain and France).

As a result of the displacement of distributionareas of many species toward the poles, the pressureon marine ecosystems increases in the polar regionsdue to the immigration of new species, while theinhabitants of these regions, adapted to cold temper-atures, cannot move to cooler latitudes. They are


therefore particularly sensitive to climate change, sothat losses of habitats and species are to be expected,especially in the polar marine sea-ice ecosystems(Smetacek and Nicol, 2005;ACIA, 2005). In addition,regional expressions of global warming are especiallyevident in the Arctic, in part because there is a par-ticularly strong feedback with regional temperaturesthere as a result of the albedo changes due to retreat-ing sea ice.

It is likely that primary production in the ArcticOcean will increase due to climate warming, albeitfrom a low initial level (ACIA, 2005). The increasedproduction can either be exploited by the zooplank-ton or fall out as sedimentation and provide nutrientsfor the benthic fauna. In regions with seasonal icecover, a temporal shift could occur between the phy-toplankton bloom and the massive occurrence ofzooplankton as well as between the zooplankton andfish larvae due to the climate-dependent changes inthe start of ice melting in spring. Such a lack of syn-chronization would result in a lower share of the pri-mary production being available for higher levels ofthe food web. Reliable predictions cannot be made,however, concerning the effect of increased primaryproduction on fish, bird and mammal populations(ACIA, 2005).

An important question is whether anthropogenicclimate change can influence naturally occurringregime shifts (Section 2.2.1). With the low resilienceof marine ecosystem structures and the intensity ofexpected anthropogenic climate signals (Section2.1.1) the possibility that regime shifts in the futurewill exhibit a different quality, occur more often orrarely, or occur in regions where they have not beenpreviously seen can absolutely not be excluded. Theobserved acceleration in the periodicity of regimeshifts in the North Pacific (King, 2005) could beindicative that there is a link with anthropogenic cli-mate change, although a conclusive judgement can-not be made at this time (Brander, 2005).

The new findings in marine ecology support theincreased climate change mitigation efforts that havebeen called for in previous WBGU reports (e.g.WBGU, 2003), because if climate warming continuesunchecked, severe unpredictable and undesirablechanges in marine ecosystems cannot be discounted.

2.3In focus: Climate and fisheries

For over 2600 million people fish is the basis of atleast 20 per cent of their protein supply (FAO, 2004).Industrial fisheries are growing and are coming intoincreasing competition with the 30 million traditionalfishers, who are often faced with losses in income

(World Bank, 2004). World fish production in recentyears has remained stagnant at around 130 milliontonnes per year, whereby the proportion of fishcaught in the sea has slightly decreased and the shareof aquaculture has risen (FAO, 2004). Fish stocks arevery unequally distributed in the sea: less than 7 percent of the ocean’s area is represented by the conti-nental shelves (water depths <200m), but theseaccount for more than 90 per cent of the global fishcatch (Pauly et al., 2002).

At the same time, the human impact on the marineecosystems of the continental shelves is especiallygreat: overfishing (FAO, 2004; MA, 2005b), includingillegal or unregulated fishing (Gianni and Simpson,2005), degradation and destruction of marine andcoastal habitats (such as corals, Section 2.4), the inva-sion of alien species, pollution of the world’s oceans(GESAMP, 2001), and, as a new threat, acidification(Chapter 4) endanger the health of the ecosystemsand the sustainability of their use. Poor fisheries man-agement and resulting overfishing are certainly morecritical factors for the fish stocks than the anthro-pogenic climate change observed so far (Worm andMyers, 2004;ACIA, 2005). In the future, however, thelatter could also cause a considerable additional bur-den for marine ecosystems (IPCC, 2001b; Richardsonand Schoeman, 2004; Section 3.1.4).

2.3.1Changes in fish populations

Similar to the terrestrial realm, the species in marineecosystems often respond to anthropogenic warmingwith a poleward shift (Parmesan and Yohe, 2003).This is also the case for many of the fish populationsin European shelf waters, with increasing evidence ofa northward shift due to warming (Fig. 2.3-1). Thestocks of cod in the North Sea are decreasing at a ratethat cannot be explained by overfishing alone.Today,the upper limit of the thermal tolerance window hasalready been reached there, with the result that pop-ulations are moving northward. The decrease of codcorrelates significantly with the changed speciesassemblage, stock decline and smaller average bodysize of the zooplankton (Beaugrand et al., 2003),which can probably be attributed to climate change.Fundamental changes in the pelagic ecosystem havebeen observed in the North Sea from 1925 to 2004,with a clear shift of many populations northward andthe immigration of southern species (Beare et al.,2004). These systematic long-term trends correlatewith the rising sea temperature. From observations inthe North Sea, Perry et al. (2005) conclude that a fur-ther increase in temperature will result in additionalchanges in the species composition and ecosystem

16

17In focus: Climate and fisheries 2.3

structure that cannot be predicted in detail, but willprobably put considerable adaptive pressure uponcommercial fisheries. In various Arctic marineregions, with a regional warming of 1–3°C, north-ward displacements of fish populations can beexpected, along with the establishment of discretepopulations (e.g., cod near Greenland) as well as theimmigration of southern species (ACIA, 2005).

2.3.2Regional prognoses of impacts on fisheries

For some marine regions, especially for waters in thenorthern latitudes, our understanding of the ecosys-tem structures and their response to natural climatevariability is good enough to discuss possible impactsof climate change. The Norwegian Sea, for example,is a very well studied area (Skjoldal, 2004). Based onexperience with natural climate variability it can beassumed that regional temperature increases of 2–4°C could increase the primary and secondary pro-duction of the sub-Arctic part of the Norwegian Seaand therefore improve conditions for fish production(Skjoldal and Sætre, 2004). At the same time, how-ever, the spectrum of species would experience a

shift, i.e. warm-water southern species would beintroduced to the area.

The authors of the Arctic Climate Impact Assess-ment (ACIA) also assume that a regional warming of1–3°C would improve the conditions for some eco-nomically important fish populations, such asAtlantic cod or herring, because the retreat of sea icewould increase both the primary and secondary pro-duction as well as allowing these species to spreadnorthward (ACIA, 2005).

Regime shifts with distinct changes in speciescomposition (Section 2.2.1) are not ruled out by theACIA, but its authors deem that the adaptation ofthe fisheries sector to the new conditions should notpresent a great expense.The ACIA comes to the gen-eral conclusion that the type and effectiveness offisheries management – in particular the preventionof overfishing through application of the precaution-ary principle – will have a greater impact on produc-tion in the Arctic than the moderate regional climatechange of 1–3°C that is projected for the 21st century.Accordingly, significant economic or social effects atthe national level are not expected, even though indi-vidual Arctic regions that are heavily dependent onfisheries could be clearly impacted.

Kara Sea

Barents Sea

Arctic Ocean

Atlantic Ocean

Cod

Herring

Mackerel,Bluefin tuna

Mackerel

North Seaherring

Anchovy, Sardine

Capelin

Nordic Seas

FFiigguurree 22..33--11Likely extension of thefeeding area for some of themain fish populations if seatemperature increases.Source: ACIA, 2005 modifiedafter Blindheim et al., 2001


This assessment changes, however, in the case of amore substantial regional climate change (>3°C).While the authors of the ACIA consider it possiblethat negative consequences could result for fisheriesin some Arctic marine regions, they hesitate to makepredictions for most Arctic regions due to the incom-plete understanding of ecosystem structure anddynamics. Although the Arctic waters have beencomparatively well studied, no reliable ecosystemmodel coupled with climate scenarios currentlyexists. The assessment of ecosystem impacts there-fore must remain speculative. In order to answer theremaining questions, research efforts will have totake a more ecosystem-based approach. Improvednumerical ecological models based on integratedenvironmental monitoring will make an importantcontribution (Skjoldal and Sætre, 2004; Section 2.7).

2.3.3Global prognoses of impacts on fisheries

The Food and Agriculture Organisation of theUnited Nations (FAO) refers to future anthro-pogenic climate change as an example of the uncer-tainty justifying a precautionary approach to fish-eries management (FAO, 2000). In its report ‘TheState of World Fisheries and Aquaculture 2002’, FAOdraws attention to the importance of natural long-term climate variability for the development of fishstocks in a chapter dedicated to this subject. It alsopoints out that global warming could have significantimpacts – positive or negative – on some, if not mostof the commercial fish stocks (FAO, 2002). It con-cludes that stocks drastically reduced by overfishingare more vulnerable to climatic changes than sus-tainably exploited stocks (FAO, 2004). However,FAO’s long-term projections are, even today, stillbased inter alia on the assumption that environmen-tal conditions, including the climate, are not changingsignificantly.

The Intergovernmental Panel on Climate Change(IPCC, 2001b) points to the increasingly acknowl-edged relationship between natural climate variabil-ity and the dynamics of fish stocks and concludes thatglobal warming complicates these relationships andwill make fisheries management more difficult. Cli-mate change therefore has the potential, during thecoming decades, to become an important factor inthe management of marine resources, although theeffects will vary widely depending on the region andecosystem characteristics (IPCC, 2001b).

The authors of the Millennium Ecosystem Assess-ment also warn about the consequences of climatechange, although they have not carried out a detailedanalysis. They describe current knowledge about the

effects of climate change on marine ecosystems asinadequate. They point out in particular that theresponse of fish stocks to environmental influencesdepends, not least, on population size. Healthy stockswith large production of fish larvae can adapt betterto population displacement and changes in ecosys-tem structure. Stocks that are greatly reduced due tooverfishing respond more sensitively to environmen-tal influences such as climate change (MA, 2005b)because there is a greater probability that the mini-mum stock level for reproduction is not attained.

Despite the lack of scientific data, several generalrecommendations for the management of marineecosystems and fisheries management can be made.These will be discussed in Section 2.6.

2.4In focus: Climate and coral reefs

Tropical coral reefs are recognized as the mostspecies-rich of marine biotopes, not so much becauseof the abundance of species of the reef-buildingcorals themselves (over 835 species have beendescribed), but because of the biological diversity oforganisms that live on and from coral reefs, repre-senting an estimated 0.5–2 million species (Reaka-Kudla, 1997). Coral reefs provide important productssuch as fish and building materials (blocks of corallimestone). They also offer protection from theeffects of tsunamis and coastal erosion, and at thesame time, because of their aesthetic and culturalvalue, they are an important source of income fromtourism. Although coral reefs only cover 1.2 per centof the global continental shelves, it is estimated thatmore than 100 million people are economicallydependent on them (Hoegh-Guldberg, 2005). A sta-tus report on worldwide coral reefs (Wilkinson, 2004)provides information on their development since the1950s and raises vital concerns with its estimation ofthe future trends:– 20 per cent of all coral reefs have been effectively

destroyed and show no immediate prospects ofrecovery,

– 24 per cent of all coral reefs are under imminentrisk of collapse through human pressures,

– a further 26 per cent are under a long-term threatof collapse.

The changes of the past 20–50 years are referred to asthe ‘coral reef crisis’ because the adaptive capacity ofcorals and the animals and plants associated withthem to changing environmental conditions has beenexceeded worldwide (Hoegh-Guldberg, 1999; Pan-dolfi et al., 2003).The pressure from human activitiesis locally generated, first through poor land manage-ment practices, whereby sediments, nutrients and

18

19In focus: Climate and coral reefs 2.4

pollutants are released and washed into the sea anddamage the reefs. In addition, overfishing, primarilythe fisheries using destructive methods (dynamite,cyanide, heavy fishing rigs), reduces the populationsof key species on the reef, damaging the function ofthe ecosystem and reducing productivity. Afterecosystem damage, macroalgae have an advantageover the coral in their growth because the feedingpressure by selectively caught fish that normally feedon these algae declines.

In addition to the local stress factors, two results ofglobal climate change are becoming increasinglyimportant to the condition of coral reefs and willtherefore be investigated in more detail in this sec-tion: the increase in seawater temperature and theacidification of seawater. These two factors con-tribute individually as well as synergistically, togetherwith the local anthropogenic stressors, to the destruc-tion of coral reefs.

It is only in recent decades that coral reefs werealso discovered to exist in deep, dark, cold-waterzones in practically all of the world’s oceans (Frei-wald et al., 2004). Their ecosystems and the seriousdangers to them, particularly from bottom-trawl fish-ing, are currently being researched.Whether they arealso threatened by the effects of climatic changessuch as temperature change and changes in the avail-ability of calcium carbonate is not clear.

2.4.1Warming impact on corals

Coral reefs dominate tropical coasts at latitudesbetween 25°N and 25°S, which corresponds to a sea-water temperature range of 18–30°C (Veron, 1986).Along with the atmosphere, the surface layers of theocean have also warmed in recent decades (Section2.1.1). In seven tropical regions where corals occur, awarming of 0.7–1.7°C has been measured in the 20thcentury (Hoegh-Guldberg, 1999).

Since 1979 a new phenomenon has been describedwith increasing frequency and geographic extent,called coral bleaching.This refers to the loss of single-celled algae that live in symbiosis with the corals. If acoral is subjected to a stress situation, which either innature or in the laboratory can be produced by highor low temperatures, intensive light, changes in salin-ity or other physical, chemical and microbial stressfactors, the algae will be expelled from the coral tis-sue.The living tissue of the corals is transparent with-out algae cells, so the white limestone skeleton willshow through – hence the term coral bleaching. Thisphenomenon is to some extent reversible becausealgal cells can be taken up again by the body tissue.

But after extended periods of coral bleaching thecorals die.

Abundant occurrences of coral bleaching werefirst described in the scientific literature in the early1980s. Strongly increasing worldwide occurrencescorrelate with higher surface temperatures of seawa-ter and with disturbances related to El Niño events(El Niño/Southern Oscillation, ENSO). The mostintense event by far occurred in 1997–1998, resultingin the death of 16 per cent of all tropical corals world-wide. Regionally the values were even higher, e.g. at46 per cent in the western Indian Ocean (Wilkinson,2004).

The strength and duration of the temperatureanomalies are important values for predicting coralbleaching. The ‘Degree Heating Weeks’ (DHW),which aggregate the thermal stress over 12 weeks,were developed as an indicator. One DHW is equalto one week with a temperature of 1°C above thesummer maximum during the previous 12 weeks.TheUSA’s National Oceanic and Atmospheric Adminis-tration (NOAA) provides an operational early warn-ing system for this. Analyses of the measurementseries show that 8 DHW led to coral bleaching in 99per cent of all cases. Coral bleaching can be predictedtoday with over 90 per cent probability several weeksbefore the event occurs (Strong et al., 2000). Theworldwide area of coral reefs affected by DHW >4 iscontinuously increasing (Wilkinson, 2004). Model-ling calculations based on IPCC scenarios indicatethat between 2030 and 2050 events similar to theanomalous year of 1998 could occur annually,spelling the end of coral-dominated ecosystems(Hoegh-Guldberg, 2005). By combining the data ofthe NOAA early warning system with global circula-tion models, Donner et al. (2005) arrived at similarconclusions. According to their research, in 30–50years coral bleaching will occur every one to twoyears in the large majority of all coral reefs if thecorals do not adapt their temperature tolerance by0.2–1°C per decade.

An important observation is that the thresholdvalue of seawater temperature for triggering coralbleaching at many locations is only 1–2°C above themaximum summer temperature. Tropical corals aretherefore living very close to the highest temperatureat which they can survive (Hoegh-Guldberg, 1999).Assuming that the near-surface seawater tempera-tures will continue to increase, the question is howcorals could respond to this temperature increase.Hughes et al. (2003) describe possible responses: asingle threshold value for all coral species is unlikely;instead, the threshold values vary within a certainbandwidth depending on coral species, water depthand location. A model in which the different thresh-old values for the demise of the corals change with


time through acclimatization and evolution seems tobe most realistic. Symbiotic algae that occur in vari-ous genotypes are adapted to different upper tem-perature limits, for example. After coral bleachinghas occurred, heat-tolerant algal groups could betaken up into the coral tissue, offering improved pro-tection against future temperature peaks, andthereby providing some limited adaptation to cli-mate change (Baker et al., 2004; Rowan, 2004).Hoegh-Guldberg (2005), however, expresses the con-cern that the evolutionary adaptation of corals andalgae cannot keep pace with the rapid environmentalchanges taking place over just a few decades.

Coral reefs could also respond to increased sea-water temperatures with a shift of their distributionor a change in their species assemblage. A polewarddisplacement of the distribution region, however,could only amount to a few degrees of latitude atmost, because both the light (for photosynthesis ofthe symbiotic algae) and the aragonite supersatura-tion (for calcification) are limiting factors (Budde-meier et al., 2004).

2.4.2Acidification impact on corals

The acidification of the sea through hydrolysis ofCO2 in seawater (Section 4.1) influences the carbon-ate chemistry and thereby also affects the corals,which produce skeletons of calcium carbonate (Orret al., 2005). The formation of limestone (calcifica-tion) is not only the foundation for the growth of thecoral reefs but also helps to counteract the process ofreef erosion.The CO2-determined impairment of cal-cification hampers the spread of coral reefs to coolermarine regions. Consequently, both increased tem-peratures and increased CO2 concentrations must beexpected to drastically constrain the distributionareas of the present-day coral reefs (Hoegh-Guld-berg, 2005).

In laboratory experiments simulating a doublingof the CO2 concentration in the atmosphere, the cal-cification rate for corals dropped by 11–37 per cent(Gattuso et al., 1999). Modelling calculations byKleypas et al. (1999) confirm these results.Accordingto their findings, calcification today has already fallenby 6–11 per cent compared to pre-industrial rates.With a doubling of CO2, a further drop of 8–17 percent compared to today’s rates was calculated.Decreased calcification results in slower expansionof the coral skeleton and therefore a decreased com-petitive capacity for space in the coral reef. In addi-tion, skeletons of lower density are produced, whichare more delicate and vulnerable to erosion.

The calcification rate is influenced not only byCO2 concentrations but also by water temperature.Increased seawater temperatures can lead to highermetabolic activity and increased photosynthesis ratesof the symbiotic algae and thus to increased calcifica-tion by the corals (Lough and Barnes, 2000). McNeilet al. (2004) conclude from in-situ investigations andmodel calculations that the calcification rates ofcorals in the year 2100 could be at as much as 35 percent above the pre-industrial rates in spite ofdecreasing aragonite saturation due to marine warm-ing, presuming an adaptation of the corals to higherseawater temperatures. These hypotheses are scien-tifically contested (Kleypas et al., 2005). For calcifica-tion to increase over the long term the temperaturerise of seawater has to remain below the thermal tol-erance limit of the corals. So the key question here iswhether the tropical corals and their symbiotic algaecan genetically adapt their temperature tolerancequickly enough to keep pace with rising seawatertemperatures.The question of possibly increased cal-cification would be moot if the corals die from heatstress.

2.4.3Measures for coral conservation

Due to the specialization of tropical coral reefs with-in a narrow range of temperatures, aragonite super-saturation, and high light availability conditions, cli-mate change, in addition to local anthropogenicstress factors, poses a great threat to them. Increasingoccurrences of coral bleaching highlight the need forrigorous implementation of climate policy measures.Even the healthiest reefs are not immune to theseimpacts, as the status report on coral reefs points out(Wilkinson, 2004). It has, however, been found that‘healthy’ reefs located in pristine areas have thegreatest chance of surviving coral bleaching episodes.So it makes good sense to strengthen the resilience ofcoral communities through protective measures.

For this purpose the establishment of marine pro-tected areas (MPAs) is considered to be especiallyeffective, preferably in their most stringent form asNo-Take Areas, which are closed to fishing (Hugheset al., 2003; Bellwood et al., 2004; Section 2.6.2). Thefocus on protected areas, however, should not lead toneglect of the remaining much larger reef areas thatare not designated as protected. The critical func-tional groups (communities of particular, oftenregionally different species that maintain the ecosys-tem) have to be protected at a regional level; other-wise, the area loses resilience.

20

21Guard rail: Conservation of marine ecosystems 2.5

2.5Guard rail: Conservation of marine ecosystems

2.5.1Recommended guard rail

The guard rail concept devised by WBGU helps tooperationalize the guiding principle of sustainabledevelopment (Box 1-1).A guard rail for conservationof marine ecosystems can be developed, although itwill inevitably be temporary in nature because thescientific basis remains weak. Analogously to theecological guard rail recommended by WBGU(2001) for terrestrial land and freshwater ecosystems,the Council recommends that at least 20–30 per centof the area of marine ecosystems should be desig-nated for inclusion in an ecologically representativeand effectively managed system of protected areas.

2.5.2Rationale and feasibility

The rationale behind this guard rail is, among otherthings, the realization that ecosystems and their bio-logical diversity are vital for the survival ofhumankind because they fulfil a great variety of func-tions and provide a whole range of products and ser-vices (MA, 2005b). Ecosystem conservation is there-fore an indispensable component of sustainabledevelopment. In its biosphere report, WBGU (2001)developed five principles that can provide a basis forsustainable management of ecosystems and serve asa background for developing a guard rail for protec-tion of marine ecosystems: (1) preserve the integrityof bioregions; (2) safeguard biological resources; (3)maintain biological potential for the future; (4) pre-serve the global natural heritage; (5) maintain theregulatory functions of the biosphere.

Protected near-natural marine ecosystems fulfilmany important functions for human society (Sec-tion 2.6).They play a major role in coastal protection(e.g. protecting coasts from sediment losses, waveerosion and flooding; Section 3.2), water purification,as a fisheries management instrument (Gell andRoberts, 2003; Section 2.6.2.1) and in tourism. Theyare also indispensable for conserving biologicaldiversity and increasing the resilience of marineecosystems to anthropogenic stress factors.

Developing a marine protected areas network bythe year 2012 has now become an internationally rec-ognized goal (Section 2.6.2.2). Although there is nodispute regarding the normative principles and thevalue or services provided by marine ecosystems, and

the need to protect them, it is very difficult to trans-late this into a quantitative guard rail because the sci-entific basis for such quantification remains weak.Moreover, a simple global ‘protection standard’ isunlikely to meet the needs of different regions withvastly diverse ecological assets and situations.A stan-dard of this sort can therefore only serve as a roughyardstick; it cannot be applied directly to all regions(Bohnsack et al., 2002;Agardy et al., 2003; Rodrigueset al., 2004). Conversely, the current practice of leav-ing almost all marine and coastal ecosystems open tooverexploitation or destruction is certainly not a sit-uation that can be considered tolerable. For this rea-son, a global guiding principle should be establishedthat helps communicate the considerable deficitsthat currently exist and make initial progress at leasttowards slowing the continuing destruction of thenatural resource base.

The IUCN World Parks Congress recommendedprotecting 20–30 per cent of each type of marinehabitat (WPC, 2003a), and in the Convention on Bio-logical Diversity this was also the target under con-sideration, although in the end it was not approved(CBD, 2003). At national level, similar coverage tar-gets are under discussion: USA: 20 per cent (NRC,2001), Great Britain: 30 per cent (Royal Commissionon Environmental Pollution, 2004); the Bahamas,Canada and the Philippines, for example: 20 per cent(Agardy et al., 2003). Australia has demonstratedthat these figures are not unrealistic by increasing theGreat Barrier Reef protected area in recent decadesfrom less than 5 per cent to 33 per cent. Due to theconsiderable uncertainties as regards scientific infor-mation, specific figures for coverage targets can onlybe temporary until better data and estimates becomeavailable.

Worldwide, significantly less than 1 per cent of themarine area is currently protected (Chape et al.,2005). In view of this fact, in addition to the need toestablish a specific coverage target, there is consider-able need for action, which is discussed in more detailin Section 2.6.2. As a basis for comparison: on land,around 12 per cent of land areas are protected(WPC, 2003b), which is much closer to the target cov-erage for ecosystem protection for terrestrial areas(10–20 per cent; WBGU, 2001). In terms of monitor-ing the implementation of coverage targets, theUNEP World Conservation Monitoring Centre andIUCN represent experienced and competent institu-tions that would be able to undertake monitoringactivities if suitably equipped. There are also report-ing obligations to be fulfilled, for example in the con-text of the Convention on Biological Diversity andthe Ramsar Convention.

With coverage targets of this sort – as in the caseof terrestrial areas – it cannot be emphasized too


often that designation of protected areas is not suffi-cient in itself to ensure that protection actually takesplace; good management and adequate funding arealso prerequisites (WBGU, 2001). In addition, theremaining 70–80 per cent of the marine area not cov-ered by protected area status must also be managedsustainably with integrated management conceptsbased on the ecosystem approach. Protected areasalone cannot stop the loss of biological diversity(WBGU, 2001), especially if overfishing is not haltedand if there is a shift in climate zones. In the case ofthe ecosystems guard rail, moreover, the principleapplies that adherence to the guard rail will only pro-vide protection for marine ecosystems if the otherguard rails too are implemented, especially the guardrails on climate protection (Box 1-1) and on oceanacidification (Section 4.4). Even the biggest and mostproficiently managed protected areas system is onlyable to mitigate the consequences of unbridled cli-mate change or extreme acidification to a very lim-ited extent: the result would be an intolerable loss ofecological services over a large area.

2.6Recommendations for action: Improving themanagement of marine ecosystems

Anthropogenic climate change has the potential tocause considerable additional stresses to marineecosystems in future (Section 2.2–2.4). It is likewisepossible that it will have an impact on commercialfishing, given that the naturally occurring variabilityin climate already plays a major role in the fluctua-tion of fish stocks. In some regions, anthropogenictemperature change is on the point of exceeding thehighest levels ever reached by natural variability (e.g.in the Arctic:ACIA, 2005). Given the current state ofknowledge, however, it is virtually impossible tomake globally aggregated forecasts of the impact ofclimate change on marine ecosystems. As no compa-rable historical data or empirical figures are avail-able, forecasts would amount to little more than spec-ulation.

Mitigation of climate change, particularly by sub-stantially reducing greenhouse gas emissions(WBGU, 2003; Schellnhuber et al., 2006), is crucial ifadditional stresses on marine ecosystems are to belimited. One mitigation option of direct relevance tooceans is sub-seabed storage of CO2 (Chapter 5).Due to the geophysical lag effects of the climate sys-tem, however, adaptation measures will be unavoid-able even if rigorous efforts are made to reduce emis-sions. For this reason, adaptation to climate changewill be the focus of this section. Priorities set byWBGU in this context are fisheries management and

marine protected areas. It makes sense to adoptadaptation measures for other reasons, too, as cli-mate change is only one of many ways in whichhuman influence degrades marine ecosystems (over-fishing, destruction and pollution of marine ecosys-tems, invasion of alien species, etc.; GESAMP, 2001;UNEP, 2002). Even considered separately, each ofthese factors poses a considerable challenge to theinternational community.

Coupling and synergistic effects between the dif-ferent factors call for particular attention (Brander,2005). A coral reef that has suffered prior damagedue to pirate fishing with poison or dynamite will beparticularly sensitive to periods of unusually hightemperatures (Section 3.3; Wilkinson, 2004). Wherestocks of a species have been heavily depleted byoverfishing, they will regenerate much more slowly ifcoastal ecosystems that serve as nursery grounds areexposed to severe stresses as a result of infrastructuremeasures or pollution, or if there is an additionalstress in the form of warming. It is easy to find otherexamples to add to this list (for an overview, seeBrander, 2005). Consequently, it is vital to considerthe various factors in an integrated manner if man-agement of marine ecosystems is to be successful. Inthe coming years, it will therefore become all themore important to rein in current overfishing prac-tices and other destructive anthropogenic factors atthe same time, so that marine ecosystems will havesufficient resilience to cope with climate change(Brander, 2005).

For these reasons, the ecosystem approach forpromoting conservation and sustainable use ofecosystems and their living resources that was devel-oped under the Convention on Biological Diversityand reaffirmed at the World Summit on SustainableDevelopment (WSSD) is vitally important (see e.g.OSPAR, 2003). In order to implement this approach,research and monitoring of marine ecosystems andocean regimes must be improved and this knowledgeapplied to the assessment and management of fishspecies of commercial interest (FAO, 2003; Section2.7). Current knowledge regarding the exceedinglycomplex interactions between climate, physical andchemical conditions in the sea, marine ecosystemsand fishery is inadequate for making reliable predic-tions relating to how marine systems are likely torespond to climate change (ACIA, 2005; Section 2.7).Inadequate knowledge must not, however, serve as apretext for delaying conservation and managementmeasures. On the contrary: in accordance with theprecautionary principle, action must be taken even ifuncertainty prevails. This precautionary principle isalready enshrined in multilateral fisheries policy, e.g.,in the United Nations agreement on migratory fishstocks.

22

2.6.1Fisheries management

National and international institutions face the chal-lenge of dealing with the complex set of anthro-pogenic factors that currently characterizes the fish-eries sector, of making decisions regarding sustain-able management of the sector on this basis and, notleast, implementing these decisions on the ground.Up to now, the situation has been less than satisfac-tory: calls for sustainable management of fish stocks,which we have been hearing for decades now and arereiterated time and again at international confer-ences, have hardly brought about any improvementin the overall situation (although there are someimportant regional exceptions) (Section 2.3). Half ofall fish stocks are fully exploited, while a quarter offish stocks have already collapsed as a result of over-fishing (FAO, 2004). Illegal and unregulated fishingon the high seas continues to be an unresolved prob-lem despite international efforts (FAO, 2001). In thefuture, this already very difficult situation will beexacerbated by climate change. In addition, newtechnologies have extended the boundaries of thefeasible ever further in fishing, for example fish find-ing using much-improved sounding technology, andreaching great depths or particular stocks using mod-ern catching methods. Nowadays, virtually no oceanhabitat is inaccessible to fishing activities.

For these reasons, management of fishing groundsbased on the ecosystem approach and on the precau-tionary principle is urgently needed in order to main-tain the resilience of marine ecosystems (Scheffer etal., 2001; Pikitch et al., 2004).The Agreement on Con-servation and Management of Straddling Fish Stocksand Highly Migratory Fish Stocks in the high seasapplies the precautionary principle. In FAO pro-grammes, too (e.g. in the Code of Conduct forResponsible Fisheries; FAO, 1995), the precautionaryprinciple and ecosystem conservation have played amajor role for some time. The EU’s strategy for pro-tection of the marine environment names the ecosys-tem approach as a key component (EU Commission,2005), although it excludes fisheries policy from thisstrategy. Moreover, the binding requirements it setsout are very vague, with the result that the strategy’seffectiveness is likely to depend largely on how it isimplemented by the Member States.

Broad-based enforcement of sustainable fisheriesmanagement is long overdue (Fujita et al., 2004).Thescientific and conceptual foundations have alreadybeen laid and have been reaffirmed repeatedly in theinternational policy arena. In many cases legislativeprovision at national and regional level is alreadyadequate. In the European Union, for example, the

Common Fisheries Policy has been endowed with alegal framework that is perfectly acceptable in envi-ronmental policy terms. It has yet to be rigorouslyimplemented, however, most notably implementa-tion of adherence to the scientifically-based recom-mendations from the International Council for theExploration of the Sea regarding catch quotas. Rapidelimination of excess fishing fleet capacity has alsoyet to take place (SRU, 2004). It is not the purpose ofthis section to discuss all the issues relating to globalfisheries management and its shortcomings. The aimis rather to formulate recommendations or reinforceexisting recommendations relating to the additionalproblem of climate change and its impact on fish-eries.• A paradigm shift away from publicly subsidized

overfishing (SRU, 2004) to a sustainable fisheriessector is long overdue. In order to achieve this,efforts must be urgently intensified to resolve theprimary problems of the marine fisheries sector,namely excess fishing fleet capacity, destructivefishing practices, excessive bycatch, inflated catchquotas, illegal or unregulated fishing in the highseas, habitat destruction in coastal ecosystems, andpollution.An increased drive to promote labellingof sustainable marine products is also urgentlyneeded. Implementation of the goals adopted bythe World Summit on Sustainable Development(WSSD) is a key yardstick in this context.

• Eliminating subsidies in the fisheries sector is aneffective means of slowing overfishing and puttingan end to it altogether in the long term. Estimatesof subsidies to the fisheries sector worldwiderange from US$15–30 thousand million annually(Milazzo, 1998; Virdin and Schorr, 2001). Thesesubsidies should be cut in order to reduce incen-tives to overexploit the marine environment. Atthe same time, public funds would be set free forinvestment in activities that include protecting themarine environment.

• Recent efforts to reduce fisheries subsidies in thecontext of the WTO are welcomed by WBGU.This relates particularly to subsidies in the OECDcountries and especially in the EU (SRU, 2004).The possibility of negative social and ecologicalconsequences arising as a result of cuts in subsi-dies, particularly in developing countries, due tothe search for new ways of earning an income oralternative ways of exploiting the natural environ-ment, must be explored and, where appropriate,taken into account. This must not, however, beallowed to hold up implementation of a swift andconsistent change in international policy on subsi-dies.

• Due to the complex interaction of many factors,both anthropogenic and natural, the integrated

23Recommendations for action: Improving the management of marine ecosystems 2.6


ecosystem approach for promoting conservationand sustainable use of ecosystems and their livingresources developed under the Convention onBiological Diversity and reaffirmed at the WSSDis vitally important. On the one hand, monitoringof ocean regimes and ecosystem parameters (e.g.indicator species) must be improved; on the other,the resulting knowledge concerning the state ofthe ecosystem must be integrated into the processof assessing and managing commercially impor-tant fish stocks (FAO, 2003).

• The precautionary principle must be rigorouslyapplied as the basis for fisheries management. Par-ticularly when forecasting fish stock dynamics andcalculating catch quotas on the basis of these,safety margins should be included to ensure thatstocks do not fall below the minimum required forreproduction and that the age structure of the fishpopulation remains healthy, even in the event of aregime shift induced by climate change (King,2005). Fisheries management must be enabled torespond to regime shifts in good time and withappropriate strategies (Polovina, 2005). Oneexample of the need to adapt in this way is the codfishery in the North Sea (Section 2.3.1).

• For short-term management (1–5 years), althoughanthropogenic climate change will have relativelylittle impact, interannual variability and climaticevents such as El Niño may trigger major effects(Barber, 2001). Assessing and forecasting thesefactors is an important area where research isneeded.

• The role of the future climate is currently largelyignored when developing management strategiesfor the medium term (5–25 years), either becauseit is seen as something that can be disregarded orbecause it is considered impossible to foresee.Since climate change can have considerableimpact on recruitment and distribution of fishstocks in the medium term, it will become neces-sary for fisheries management to take these effectsinto consideration. At present, the impact of cli-mate variability and climatic events on fish stockscan only be analysed after the event. In view of thefact that climate change is already apparent, fore-casting capacity needs to be developed in futureand used to conduct risk analyses.This applies par-ticularly to sensitive fish populations on the edgeof their natural distribution area.

• When developing the models that are used as thebasis for setting quotas, there needs to be a shiftaway from analysing and modelling individual fishpopulations of commercial interest to ecosystem-based models that take into account the dynamicinteractions between climate, ocean and marineecosystems (Pikitch et al., 2004).The usefulness of

static concepts based on the assumption ofunchanging environmental conditions is becomingincreasingly questionable.

• In the case of terrestrial ecosystems, subdividingthe area in question into zones with varying inten-sities of use is a long-established procedure forsolving land-use conflicts (WBGU, 2001). For theoceans too, in the context of marine spatial plan-ning systems, zoning is increasingly recognized asa useful instrument for sustainable, ecosystem-based fisheries management (Pauly et al., 2002;SRU, 2004; Pikitch et al., 2004; Boersma et al.,2004). Marine protected areas have a special roleto play as a component of marine spatial planningin this context because, in conjunction with othermeasures, they represent an important tool forimplementing the ecosystem approach. Recom-mendations relating to this are discussed in moredetail in the next Section 2.6.2.

2.6.2Marine protected areas

2.6.2.1Definition and motivation

Climate change, ocean acidification and sea-level risewill have considerable impact on the marine envi-ronment (Sections 2.2–2.4). These ‘new’ anthro-pogenic factors, moreover, are affecting marineecosystems which, in many regions, have alreadybeen significantly weakened by overfishing, contam-ination, invasive species and other human-inducedinfluences. Recommendations for improving fish-eries management were presented in Section 2.6.1above. This section will deal with marine protectedareas (MPAs), which – like their counterparts on land– are one of the most important instruments avail-able for ecosystem protection (IUCN, 1994; Kelleher,1999; Murray et al., 1999).

IUCN defines marine protected areas as ‘any areaof the intertidal or subtidal terrain, together with itsoverlying water and associated flora, fauna, historicaland cultural features, which has been reserved by lawor other effective means to protect part or all of theenclosed environment’ (IUCN, 1988).

MPAs play a particular role in protecting themarine environment, as they are a direct means ofimplementing the ecosystem approach, and one ofthe easiest to apply (Royal Commission on Environ-mental Pollution, 2004). Although they cannot halteither climate change or acidification, nor altogetherprevent invasion by alien or highly migratory species,they are an important tool for enhancing the

24

resilience and adaptive capacity of ecosystems. Theyare also important because they allow mitigation ofanthropogenic factors such as overfishing or habitatdestruction within their boundaries by means ofmanagement or prohibition (e.g. Mumby et al., 2006).MPAs are thus the most important means, for exam-ple, of dealing with coral bleaching (Section 3.3.4),because although they do not tackle the underlyingcause, they can enhance the general resilience of thereef (Grimsditch and Salm, 2005). However,enhanced scientific understanding of the relationshipbetween resilience, anthropogenic influence and bio-logical diversity is needed (Section 2.7). For coastalprotection, near-natural ecosystems are also impor-tant: for example, the Asian tsunami of 26 December2004 was able to penetrate much further inland inplaces where the mangroves or coral reefs had beendestroyed than elsewhere (Danielsen et al., 2005;Fernando and McCulley, 2005). Protected coastalecosystems are therefore also an important compo-nent of strategies for adapting to climate change(Section 3.4.1).

In addition to their role in ecosystem protection,MPAs can also be useful as a fishery managementtool for conservation of commercial fish stocks, e.g.where traditional management has failed and over-fishing has resulted, or to safeguard against futuremistakes of this sort (Bohnsack, 1998; Pauly, et al.,2002; Gell and Roberts, 2003). Even conservation-based fishery can have a range of negative effects onmarine ecosystems, and establishing MPAs can miti-gate these (Palumbi, 2003). MPAs can also provide aretreat for species that are fished, but are not subjectto monitoring or management. Coastal ecosystemsand estuaries are also important for protecting thenursery grounds of many fish species against climatevariability (Attrill and Power, 2002). MPAs should beviewed in conjunction with the traditional tools offishery management, for one reason because quotasetting could also be affected by the establishment ofa large-scale network of MPAs if fishery activities arerestricted to the areas outside MPAs (Hilborn, 2003).

A graded system of protected area managementcategories is applied in the marine environment too.This ranges from total protection (marine reserveswhere extractive use is prohibited) to areas servingprimarily to uphold sustainable and/or traditionaluse of marine resources (IUCN, 1994).Areas that areclosed to fishing activities (no-take areas) representa special type of MPA. Different categories of pro-tected area often sit side by side, with core areasunder strict protection and peripheral zones withfewer restrictions relating to use (Agardy et al.,2003). The effectiveness of MPAs can be improved ifthey form part of a protected areas system geared

towards ensuring ecological representativeness andcreating networks.

Although differences of opinion still persist withregard to the optimum design and management ofmarine protected areas (NRC, 2001), there is broadconsensus that adaptive management, linking of indi-vidual MPAs to protected areas systems, participa-tion or co-management and an integrated view of therelationship between MPAs and the intensivelyexploited areas outside them are important aspectsof MPA design and management.

2.6.2.2International policy objectives

Because of the double use of MPAs for ecosystemconservation on the one hand and as a fishery man-agement tool on the other (Lubchenko et al., 2003),as a guard rail, WBGU recommends designating20–30 per cent of the marine area for inclusion in alinked system of MPAs (Section 2.5.1). Current pro-tected areas coverage amounts to less than 1 percent of marine habitats. Considerable catching up istherefore required and it is only very recently thatthis has led to the formulation of a number of policyobjectives in this area:• At the WSSD, the international community set

itself the target of establishing an ecologically rep-resentative and effectively managed network ofmarine protected areas by 2012 (WSSD, 2002).

• The World Parks Congress reaffirmed this goal in2003 and made it more specific with the recom-mendation that at least 20–30 per cent of everymarine habitat should be strictly protected (WPC,2003a).

• In the context of its programme of work on pro-tected areas, the Convention on Biological Diver-sity has adopted the WSSD target for protectedareas, albeit without specifying a coverage per-centage (CBD, 2004a).

• A regional example is the OSPAR/HELCOMConvention, which has also set itself the target ofcreating a well managed and ecologically coherentsystem of marine protected areas by the year 2010(OSPAR, 2003).

2.6.2.3Present international law

Although the concept of MPAs or related strategiesare used in the different international conventions,they are not used consistently and thus need to beexpressed in more concrete terms (Agardy et al.,



2003). Policy objectives are set out in the followingprovisions of international law:• The Convention on Biological Diversity – whose

objectives also cover marine ecosystems – envis-ages protected areas as an in situ conservationmeasure (Art. 8 a).

• The United Nations Convention on the Law of theSea (UNCLOS) makes explicit mention of specialprotected areas only in the following context: Art.211, para. 6 allows tightening of protective regula-tions in clearly defined areas in connection withmeasures to prevent pollution of the marine envi-ronment from vessels. Coastal states wishing toapply this provision can designate such an areawithin their respective exclusive economic zone,and apply to the competent international organi-zation, the International Maritime Organization(IMO), requesting affirmation of the need for spe-cial measures to protect this area. Reasons may berelated to the area’s oceanographical and ecologi-cal conditions, its utilization or protection of itsresources. Under this provision, however, specialmeasures are restricted to regulations for the pre-vention of pollution from vessels.

• In various agreements in the field of internationallaw on protection of the marine environment,there are terms and concepts that have similarobjectives to MPAs.The International Conventionfor the Prevention of Pollution from Ships (MAR-POL Convention), for example, provides for theestablishment of ‘Particularly Sensitive Sea Areas’(PSSAs). The purpose of the protected areas inthis case is to ensure protection from pollutionfrom ships in particularly vulnerable areas that areto be designated accordingly. Further examplesare the Convention for the Protection of theMarine Environment of the North-East Atlantic(OSPAR Convention) and the Convention for theProtection of the Mediterranean Sea against Pol-lution (Barcelona Convention), both of whichhave a specific additional annex (OSPAR) or pro-tocol (Barcelona Convention) on the establish-ment of ‘Specially Protected Areas’. IMO hasadopted a Resolution (A.885 (21)) setting out pro-cedures for the establishment of PSSAs(Hohmann, 2001).

Requirements for designating protected areas varyaccording to the maritime area in question. The lim-its set out under international law depend primarilyon the maritime area in which the protected area is tobe established in a given case (Box 2.6-1; Proelß,2004).• In principle, in the case of internal waters and ter-

ritorial sea, the coastal state has the freedom todecide on designation of MPAs. This may be lim-

ited to a certain extent, but only with regard torestrictions on shipping activities (Box 2.6-1).

• The legal situation with regard to exclusive eco-nomic zones (EEZ) is similar. UNCLOS accordsparticular sovereign rights to coastal states in thismaritime area concerning exploitation and con-servation of the living and non-living naturalresources in the maritime area in question, includ-ing the seabed and its subsoil. In relation to estab-lishing MPAs, this means that the coastal state hasthe freedom to adopt measures so long as thesemeasures are aimed at restricting exploitation ofnatural resources. However, the establishment ofan MPA in this area may not, for example, restrictthe right of innocent passage of foreign vessels(Box 2.6-1).

• On the high seas, although the establishment ofMPAs is not ruled out in principle (Proelß, 2004),it entails certain legal problems (Platzöder 2001;Warner 2001). These are discussed in Section2.6.2.4.

• In contrast to agreements that are limited to par-ticular regions (e.g. the OSPAR Convention, theConvention for the Protection of the Marine Envi-ronment of the Baltic Sea – the Helsinki Conven-tion, or the Barcelona Convention) there is cur-rently no global instrument of international lawthat specially promotes designation of cross-bor-der MPAs or places any obligation on states to doso.

2.6.2.4Marine protected areas in the high seas

There are significant deficits in the legislation per-taining to establishment of marine protected areas inthe high seas (CBD, 2005a). The regional multi-func-tional maritime conventions only cover a very lim-ited range of marine areas outside national jurisdic-tion, with the result that large areas of the world’soceans are not covered. In addition, existing regionalfishery management regimes are limited to particularfished species such as tuna, while species that are notintensively fished are excluded. Application of theecosystem approach under these regimes is also inad-equate.

The United Nations Convention on the Law of theSea (UNCLOS) reaffirms the right to freedom ofnavigation, which forms part of customary interna-tional law (i.e. it is a right that, in principle, cannot berestricted) (Art. 87 UNCLOS). It is thus out of thequestion to establish an MPA in the high seas thatentails the prohibition or restriction of shippingactivities. In addition, states may not conclude agree-ments establishing MPAs in the high seas that might

26


BBooxx 22..66--11

MMaarriittiimmee zzoonneess uunnddeerr iinntteerrnnaattiioonnaall llaaww

The United Nations Convention on the Law of the Sea(UNCLOS) lays down the fundamental provisions underinternational law with respect to delimitation of maritimezones, and this is also relevant for designation of MPAs. Inbrief, the maritime zones under UNCLOS are organized asfollows (Figure 2.6-1):• High seas: According to Articles 86 and 89 UNCLOS,

‘high seas’ are those parts of the sea that are not subjectto the sovereignty or jurisdiction of any state and as suchconstitute ‘an area under international administration’.The principle of freedom of the high seas applies in thisarea. This comprises primarily freedom of navigationand of overflight, freedom to lay submarine cables andpipelines, freedom to construct artificial islands andother installations, freedom of fishing and freedom ofscientific research. These freedoms may be exercised byall states, including land-locked states. In the area consti-tuting the high seas, no state may validly purport toimpose restrictions of any sort on other states relating touse of the high seas. Furthermore, international agree-ments between individual states can always only bind thestates that are party to the agreement, and not thirdstates.

A distinction is drawn between the high seas and maritimeareas that are subject to varying degrees of territorial juris-diction by the coastal state. According to Article 86

UNCLOS, the maritime areas over which the coastal statehas varying degrees of jurisdiction are:• Exclusive economic zone: Here, in the zone that lies on

the boundary with the high seas, the coastal state beginsto have the right to exercise jurisdiction based on its ter-ritorial rights. The corresponding rights of sovereignty,however, are limited insofar as they relate exclusively toexploitation and conservation of the living and non-liv-ing natural resources in the zone in question, includingthose in the seabed and its subsoil.

• Continental shelf: This term too is defined in terms of theright to exploit natural resources, although in this case itis specifically those of the seabed and subsoil close to thecoast. Thus, according to Article 77 para. 1 and 2 UNC-LOS, the coastal state exercises over the continentalshelf exclusive sovereign rights for the purpose ofexploring it and exploiting its natural resources.The def-inition of resources that applies in the case of exploita-tion of the continental shelf, however, is somewhatrestricted by comparison with that of the exclusive eco-nomic zone (non-living resources of the seabed and itssubsoil and ‘immobile’ organisms).

• Territorial sea: Here, the special rights of coastal statesare no longer limited to exploitation of marineresources, but are identical to actual territorial sover-eignty.

• Internal waters: This is where the sovereign rights andjurisdiction of a coastal state are most extensive; thismaritime zone forms an integral part of a state’s terri-tory.

InternalWaters(landwardof low-watermark)

Territorial Sea (0-12 NM)No high seafreedoms, exceptinnocent passagefor foreign ships

Contiguous Zone(0-24 NM) Control for customs,fiscal, immigration and quarantinepurposes

Exclusive Economic Zone (EEZ)(12-200 NM from baseline) Sovereignrights for exploration, exploitation, conservationand management of natural resources

High Seas Freedoms(From 12 NM seaward) High seas freedoms of navigation,overflight, laying of submarine cables and pipelines, other lawful uses by any state

High SeasTraditional high seas freedoms,including taking „living marineresources“ for fisheries anduses other than exploitation ofnon-living resources on or underthe seabed

High seas freedoms includetaking of living resources on

the seafloor

Subsoil Beneath Continental Shelf

Sea Floor

ContinentalShelf Extension(up to 350 NMfrom baseline ifapproved)

„The Area“ Pursuant to Art. 1para 1 (1) UNCLOS:Internationalmanagement fornon-living resources

Low waterbaseline 12 NM 24 NM 200 NM

FFiigguurree 22..66--11Maritime zones under the United Nations Convention on the Law of the Sea (UNCLOS). NM = nautical mile (1 NM =1.852km).Source: Gorina-Ysern et al., 2004


be detrimental to third states that are not parties tothe agreement. A corresponding commitment agree-ment among states in a particular region that are themain users of the high seas thus has no binding effecton third states. ‘Freedom of the high seas’ alsoincludes, for example, the fundamental right of everystate to use the marine resources of the high seas (e.g.fishing). In contrast to the case of freedom of naviga-tion, however, this right is not unrestricted, and cor-respondingly there is a range of international con-ventions regulating the use of living marine resourcesin the high seas, especially relating to particularspecies. Examples include the prohibition on fishingof anadromous species (e.g. salmon, which spawns infreshwater but lives in seawater) in the high seas inaccordance with Art. 66, para. 3(a) UNCLOS, or thewhale sanctuaries provided for under the Interna-tional Convention for the Regulation of Whaling(Gerber et al., 2005).

Urgent problems such as the increasing destruc-tion due to fishery activities of sensitive underseastructures that are particularly rich in biologicaldiversity (e.g. seamounts or cold-water coral reefs;UNGA, 2004; CBD, 2004b), and the scale of illegaland unregulated fishing (FAO, 2001), necessitaterapid identification and implementation of solutionsfor marine protection in the high seas. In view of theclear will of the international community to step upuse of marine protected areas as a tool, action isneeded to improve provisions pertaining to MPAs inthe high seas under international law. In developinga regime for marine protected areas in the high seas,the following specific requirements should be met(CBD, 2005a):• Moving beyond approaches that focus on specific

species or regions, the aim should be to arrive at anintegrated approach enabling the creation oflarge-scale networks for protecting marine ecosys-tems in the high seas too. Freedom of access tomarine protected areas in the high seas shouldalso be guaranteed for scientific research, insofaras this does not run counter to conservation objec-tives.

• In view of the problem of illegal and unregulatedfishing in the high seas – which cannot be tackledby individual states because they do not have ter-ritorial jurisdiction to enforce the law in this mar-itime zone – mechanisms for enforcing the rele-vant conservation requirements in the high seasmust be considered (Platzöder, 2001; Warner,2001).

• In view of the need to create networks that covera broad area, efforts should be made to ensurethat the establishment of MPAs in the high seas –contrary to what has happened hitherto under the

various specific conventions – takes place in acoordinated fashion (CBD, 2005b).

2.6.2.5Negotiation processes

At global level, negotiations on MPAs are takingplace notably in two parallel political processes:• In the Convention on Biological Diversity, MPAs

are on the agenda of a Working Group on Pro-tected Areas, including protected areas beyondthe limits of national jurisdiction. However,attempts to agree specific areas of the high seasthat might be suitable for designation as MPAs orset a specific target of establishing 5–10 MPAs inthe high seas by 2008 have so far failed due toresistance on the part of a few fishing nations (e.g.Iceland, Norway, New Zealand).

• In 2004 an informal Working Group of the Gen-eral Assembly of the United Nations was estab-lished (UNGA, 2004) with a broad mandate relat-ing to marine biodiversity conservation beyondareas of national jurisdiction.This Working Groupconvened for the first time in February 2006.Although the positions of the different countrygroups on MPAs still diverge widely, many statesacknowledge the need to act to close this gap ininternational law.

2.6.2.6Recommendations for action relating to marineprotected areas

Despite the importance of marine protected areas,protection of the marine environment must not bereduced to this one instrument alone. Adherence tothe ecological guard rail (20–30 per cent of marineecosystem areas under protection; Section 2.5) isindispensable for conservation of the marine envi-ronment, but areas outside MPAs must also be man-aged sustainably using the ecosystem approach. Aparticularly important precondition for the successof MPAs is the urgently needed enforcement of sus-tainable fisheries management (Section 2.6.1).Another precondition is adherence to the guard railson climate change and ocean acidification (Box 7-1),without which even an excellent protected areas sys-tem will lose most of its impact. In addition, it is notsufficient to plan, designate and link MPAs; theymust also be well managed and adequately funded.Only in this way is there a chance that the target ofsignificantly reducing the rate of loss of biologicaldiversity by 2010 can be met in marine environmentstoo.

28

Implementing international targets• The rate of growth of MPA coverage, currently

3–5 per cent per year, is much too low to meet theinternationally agreed targets on time, given thatcurrent coverage is less than one per cent (Woodet al., 2005). Efforts in this regard need to be inten-sified significantly.

• MPAs should be sufficiently large and linked withone another in protected area systems; theyshould encompass zones with different forms andintensities of use and be part of an integrated sys-tem of management that includes neighbouringcontinental shelf and coastal areas. In addition,they should be designed to be flexible and adap-tive, since climate change may impose the need forreorganization if ecosystem processes change orshift in location (Soto, 2002). Improving the scien-tific basis in this context should be carried out inparallel with ongoing management activities.Adaptive management strategies and flexibilityare crucial to deal with local impacts of climatechange that are difficult to forecast.

• In the territorial sea and in EEZs, states can beginto implement international objectives now with-out coming up against problems relating to inter-national law.The EU Habitats Directive and BirdsDirective are both fully applicable in EEZs. InGermany this has already taken place: in the con-text of the Natura 2000 network, around 30 percent of the German maritime area of the EEZ hasbeen registered as protected areas with the Euro-pean Commission. However, imposing restrictionson fishing is not a straightforward matter at pre-sent, as this is an area in which EU competencesapply. MPAs should be used more effectively as atool for sustainable fisheries, for example by set-ting long-term or temporary limits on fishingactivities in protected areas (SRU, 2004).

• In developing countries particularly, there is agreat deal of catching up to be done. In these coun-tries, not only are there very few MPAs, but thedesignated protected areas in many cases amountto little more than ‘paper parks’ in which effectiveprotection is not or cannot be enforced. Develop-ment cooperation should therefore make designa-tion and management of MPAs a priority. In doingso, cooperation should take place between pro-tected areas specialists and fishery representa-tives, and the local population should be involvedin planning and management.

Securing fundingDue to the difference between the guard rail for pro-tection of marine ecosystems (20–30 per cent cover-age) and current coverage (less than 1 per cent ofmarine ecosystems are currently protected; Section

2.5.2), the need for additional funding is consider-able. According to the findings of Balmford et al.(2004), the annual costs associated with protectingmarine ecosystems on this scale would be in theregion of US$5–19 thousand million. This figureincludes both ongoing costs and one-off costs relat-ing to implementation. Indirect costs, however, suchas costs to businesses in the fisheries sector arising asa result of fishing exclusion, are not included.• WBGU considers it to be the responsibility of

national governments and the international donorcommunity to ensure adequate funding for man-agement of marine protected areas. Up to now, ithas often proved impossible to ensure availabilityof adequate long-term financing: payments frompublic donors are often small and there are com-peting demands on these resources. The sameapplies to international transfer payments such asthose made by the Global Environment Facility(GEF) or by donors in the context of bilateraldevelopment cooperation (OECD, 2002; GEF,2005a).WBGU calls upon public donors to under-take additional efforts to make adequate fundingavailable on a sustained basis. Complementaryfunding may be raised by means of instrumentssuch as user charges or promoting private dona-tions for conservation measures (Emerton, 1999;Morling, 2004).

Protected areas in the high seas: Closinggaps in international lawThe ongoing process of political negotiations todevelop an instrument for designating and managingprotected areas in the high seas is to be welcomed,and the German Federal Government should givethis process vigorous support. The basis for this isUNCLOS. Despite the fact that UNCLOS laysgreater emphasis on rules pertaining to usage than onprotection and conservation of marine resources, itnevertheless also provides the primary legal frame-work for protection of the marine environment(Platzöder, 2001). Fundamentally changing UNC-LOS is not a political option, whereas adding moder-ate supplementary provisions to the current law ofthe sea seems feasible in both political and legislativeterms. Options for doing this include the following:• The primary task is to develop a multilateral

agreement on designation of protected areas andcorresponding systems in the high seas andappend this to UNCLOS, either as an additionalprotocol or as a supplementary convention. Aprecedent already exists for this type of procedurein the form of the Agreement relating to the Con-servation and Management of Straddling FishStocks and Highly Migratory Fish Stocks that istied to UNCLOS, and which is concerned primar-



ily with the exploitation of relevant fish species inareas outside national jusrisdiction.

• It would make sense to assign the monitoring andcoordination tasks to the same – yet to be estab-lished – international regime. From a legal per-spective, the mechanism should be laid down pri-marily in the above-mentioned multilateral agree-ment.With regard to this, the proposal to establisha Global Oceans Commission, put forward at thefirst International Marine Protected Areas Con-gress in 2005 in Geelong (Australia), can be con-sidered as being along the right lines.

• The Convention on Biological Diversity (CBD)should also be amended or supplemented accord-ingly, whereby care will need to be taken to avoidoverlap. The CBD has developed considerableexpertise in the field of biodiversity conservation,and should therefore be involved in the UNCLOSnegotiation process, for example by providingexpert input. At the same time, the new regime’sfunctions and areas of responsibility should beexplicitly recognized by the CBD. In order toachieve this, the relevant CBD negotiationprocesses should be intensified with the aim ofsecuring a key role for the CBD in the design ofMPAs in the high seas in terms of content, e.g.deciding on appropriate instruments and criteriafor the selection of MPAs. With a view to support-ing cross-border conservation efforts, it would beuseful to explore whether it would be worthwhiledeveloping a protocol to the CBD on protectedareas in the medium term based on the findings ofthe current Working Group. Such a protocolshould cover the whole spectrum of protectedareas and not merely MPAs.

• The informal Working Group of the GeneralAssembly of the United Nations on marine biodi-versity in areas beyond national jurisdiction hastaken the first step towards closing the gap ininternational law regarding MPAs in the high seas.At the next UN General Assembly, the GermanFederal Government should urge that this soundbasis be used to ensure continuation of the negoti-ation process.

2.7Research recommendations

Research into climatic factors• Behaviour of sea ice: Possibilities for monitoring

changes in the thickness of Arctic sea ice in partic-ular remain inadequate, and sea ice simulationmodels need to be developed further in order toimprove estimation of future sea ice dynamics.

• Behaviour of continental ice: The future behaviourof Greenland’s continental ice sheet is likely to bedecisive for the future dynamics of ocean currentsin the Atlantic. Methods of modelling continentalice sheet dynamics need to be improved signifi-cantly.

• Stability of Atlantic circulation and risks of changesin currents: Climate models continue to differ con-siderably as regards the information they provideon the future stability of ocean currents. Reasonsfor this include internal oceanic processes (such asmixing), which as yet are poorly understood, andinteractions with other climate components (e.g.the freshwater budget of the North Atlantic),which are difficult to quantify. Observation andenhanced modelling efforts could help to reducethe uncertainties in this area.

Research into marine ecosystems, fisheriesand marine protected areas • Monitoring: Observation data covering large

areas of the oceans over long periods are vital,particularly as regards the nutrient situation andplankton (especially zooplankton). Because theyprovide, among other things, important inputs formodelling marine ecosystems, support should begiven to appropriate monitoring programmes (e.g.using the Continuous Plankton Recorder).

• Understanding the systems involved: Too little isknown about the structure and dynamics ofmarine ecosystems to be able to make reliableestimates of the impact of climate change. Exam-ples include the significance of temperatureeffects on primary production, the impact of sea-ice retreat, or decoupling of trophic levels as aresult of disparities in species’ responses to cli-mate change (e.g. migration, adaptation).Increased emphasis should be given to ecosystem-based research approaches with a view to enhanc-ing understanding of the relationships betweenanthropogenic disturbance, biological diversityand resilience of marine ecosystems and incorpo-rating this into new ecosystem models. The inter-national research project GLOBEC and the newIMBER project have drawn up a detailed cata-logue of issues (GLOBEC, 1999; IMBER, 2005).Promotion of these interdisciplinary research ini-tiatives by national research promotion agenciesshould be stepped up.

• Modelling marine ecosystems: In order to gain abetter understanding of the impact of changes inclimatic factors (temperature, wind and ocean cur-rent patterns, etc.) on marine ecosystems, knowl-edge regarding the various ecosystem componentsmust be integrated into improved ecosystem mod-els and coupled with recent climate/ocean models.

30

• Improving the basis of fisheries management: Inorder to implement the ecosystem approach infisheries management, there needs to be a shift infocus in terms of model development away fromexamination of individual fish species under theassumption of constant environmental conditionsto a more integrated ecosystem modelling ap-proach. Qualitative models should also be used toachieve this, incorporating expert knowledgeregarding processes in dynamic systems (Kropp etal., 2005). Special attention should be paid to theimpact of natural climatic variability and anthro-pogenic climate change on fish population dynam-ics, as well as to the socio-economic consequencesand possible adaptation measures.

• Design and management of marine protected areas:Development of the theoretical basis for thedesign of marine protected areas should moveaway from the study of individual species towardsmulti-species, ecosystem approaches. Of particularimportance in this context is the networking ofMPAs with one another and with concepts for sus-tainable use of the surrounding coastal and marineareas. There are many unanswered questionsregarding the design of MPAs in view of climatechange and the potential for adaptation. In linewith the principle of adaptive management,research and monitoring issues need to be givengreater consideration in the design and manage-ment of MPAs. The basis for defining guard railsand coverage targets needs to be improved, par-ticularly the basis for designating the proportionof an area that should be strictly protected (no-take areas). Furthermore, there is a need toincrease research evaluating participatoryapproaches (e.g. community-based management)and the use of traditional knowledge, and to eval-uate approaches that draw on management expe-rience within the local population.

31Research recommendations 2.7

Sea-level rise, hurricanes and coastal threats

3.1Climatic factors

3.1.1Sea-level rise

3.1.1.1Lessons from Earth’s history

A rise in sea level is one of the unavoidable physicalconsequences of global warming. A close linkbetween temperature and sea level is also evident inclimate history. At the peak of the last ice age(around 20,000 years ago) sea level was around 120mlower than today, and the climate was about 4–7°Ccolder. By contrast, during the last warm period, theEem (120,000 years ago), the climate was slightlywarmer than today (by approx. 1°C), but sea levelwas probably several metres higher – estimates varyfrom 2 to 6m (Oppenheimer and Alley, 2004). Goingback farther into the Earth’s history, one can findeven warmer climate epochs. Three million yearsago, during the Pliocene, the average climate wasabout 2–3°C warmer than today and sea level was25–35m higher (Dowsett et al., 1994).

The main reason for these large sea-level changesis the change in quantities of water that are tied upon the land in the form of ice. The ‘sea-level equiva-lent’ of the ice mass on Greenland equates to 7m, theWest Antarctic ice sheet to 6m, and the East Antarc-tic ice sheet to more than 50m. Around 35 millionyears ago (in the Eocene) was the last time ourplanet was completely free of polar ice caps, thanksto high CO2 concentrations related to the plate-tec-tonic situation at the time, and sea level was almost70m higher than today (Zachos et al., 2001; Barrett,2003). In this kind of time frame, however, volumechanges in the ocean basins due to plate tectonics canalso contribute to sea-level changes.

Plotting the values above on a graph (Fig. 3.1-1)reveals a relationship between temperature and sea

level, where a global warming of 3°C corresponds toa sea-level rise of several tens of metres. This is anorder of magnitude more than the IPCC expects bythe year 2100 (9–88cm; IPCC, 2001a). The main rea-son for this apparent discrepancy is that the relation-ship shown in the figure is based on a climate nearequilibrium (following many millennia with rela-tively constant temperatures) – not during rapidchanges as they are now occurring.The numbers givea general idea of how sea level would change aftermillennia with a 3 °C warming. But they do not allowany conclusions about how fast the ice masses couldmelt with warming and how quickly sea level couldrise in response.

The end of the last ice age provides informationabout the possible rate of sea-level rise. At that timethe global average temperature rose by around 4–7 °C, an amount that is also reached in pessimisticscenarios for the future. But the warming at that timetook around 5000 years, which is much slower thanthe present trend. From 15,000 to 10,000 years agosea level rose by around 80m, an average of 1.6m percentury (Fairbanks, 1989). During some intervals

5

Pliocene 3 million years ago

100

50

-50

-100

-15010 15 20

Global mean temperature [°C]

Sea

leve

l [m

]

Today

Eocene 40 million years ago

Projectionfor 2100(+ 1 m)

Last GlacialMaximum 20,000years ago

0

FFiigguurree 33..11--11Mean global temperature and sea level (relative to today’s)at different times in Earth’s history, with the projection forthe year 2100 (1m above today’s sea level). For the long terma much higher sea-level rise probably has to be assumed thanthat predicted for 2100.Source: after Archer, 2006

3

3 Sea-level rise, hurricanes and coastal threats

rates of up to 5m per century were reached (Clark etal., 2004).

These values cannot simply be applied to today’ssituation. The ice sheets at that time were consider-ably larger, which means the melting regions on themargins were greater, allowing a greater flow of melt-water. In addition, due to Earth’s orbital cyclesaround the sun (Milankovich cycles; Ruddiman,2000), the incoming solar radiation at high latitudesof the Northern Hemisphere was considerablystronger, a situation that cannot be directly com-pared with the global increase in greenhouse gas con-centrations.These two factors suggest higher meltingrates at the end of the ice age than during the presentwarming. The much slower warming at that time, bycontrast, would suggest lower melting rates. In fact,the disappearance of ice sheets at that time for themost part kept pace with the gradual climate warm-ing, so the assumption that ice masses would havemelted significantly more rapidly with faster warm-ing is quite plausible.

Two conclusions can be drawn from this discus-sion. Firstly, rates for sea-level rise of up to 5m percentury are documented, and these probably do notrepresent an upper limit. Thus climate history showsthat a much more rapid rise than that expected by theIPCC for the 21st century is possible. Secondly, suchrates of sea-level rise suggest dynamic meltingprocesses of the ice sheets, also taking account of theconditions at the end of the last ice age. This meansthere can be not only a simple melting through con-tact with warmer air, but also an accelerated flow ofthe ice into the sea.

3.1.1.2Dynamics of the continental ice masses

The Earth presently has two large continental icesheets with a thickness of 3–4km, in Greenland andAntarctica. Both are in a steady-state: in the centrenew ice is continuously formed by snowfall, while iceflows away on the margins. Under persistently con-stant climatic conditions these processes are in bal-ance and the size of the ice mass does not change. Butin the Antarctic it is significantly colder than inGreenland. In Greenland, therefore, a large part ofthe ice at the margins melts while still on the land(like on a mountain glacier), while in the Antarctic itreaches the sea and tongues of the ice float on thewater to form ice shelves.

It is still difficult to reliably measure changes inthe total volume of these two ice masses. Effortsinclude elevation profiles taken from satellites andaeroplanes.There is still controversy over the marginof error of these measurements; they do not accu-

rately record the craggy topography often found onthe margins of the ice sheets. Newer techniquesinclude satellite measurements of anomalies in thegravitational field. Changes at the margins of the icemasses are best obtained by local measurements anddetermination of the flow rate of the ice by satellites.

The various measurement methods provide thefollowing qualitative picture for both ice sheets: inthe past ten to twenty years, the thickness in the cen-tre seems to be increasing somewhat, as should beexpected with climate warming because of increasedsnowfall. On the other hand, increasing dynamicmelting processes can be observed on the margins.The quantitative net balance of these processes is notexactly known, so a short discussion of the currentmeasurement results follows.

In Greenland around half of the ice flows out ofonly 12 fast-moving outlet glaciers; the mass balanceof the ice depends largely on changes in these iceflows (Dowdeswell, 2006). New data show that theflow rates of many of these glaciers (among othersthe Jakobshavn Isbrae) have doubled in recent years(Joughin et al., 2004; Rignot and Kanagaratnam,2006). Furthermore, measurements of the melt area,which can be determined from satellite pictures,show an increase of around 25 per cent from 1979 to2005 (Fig. 3.1-2); the area reached its highest extentever in the year 2005 (Steffen and Huff, 2005). Whenthe area that is affected by melting increases, itshould cause a loss of mass in the ice cap. It has alsobeen found that meltwater from the ice surface runsthrough holes (so-called glacier mills) to the base ofthe ice and acts like a lubricant there, accelerating theflow of the ice (Zwally et al., 2002).

Rignot and Kanagaratnam (2006) conclude thatthe acceleration of the ice flow represents a loss ofmass corresponding to 0.5mm of sea-level rise peryear, and that this value has doubled in the past tenyears. This is equal to one-sixth of the current mea-sured global sea-level rise (Fig. 3.1-4). This is in con-trast to measurements of the elevation of the ice bysatellite altimeters (Johanessen et al., 2005; Zwally etal., 2005), which indicate an increase in the mass ofthe Greenland ice (corresponding to a sea-levelchange of -0.03mm per year), but which do not accu-rately register the small-scale processes at the mar-gins. Because this increase is significantly smallerthan the loss observed by Rignot and Kanagaratnam,a net mass loss of Greenland ice has to be assumed,although there are considerable uncertainties in thenumbers, and the various measurement methods yetneed to be better reconciled.

More important, however, than the presentchanges in mass balance, which are still small andimpossible to record accurately, is what is to beexpected in the future with progressive warming.

34


Model calculations show that with a warming of thenear-surface air layer above Greenland of about 2.7°C or more, it is likely that the entire ice sheet willgradually melt (Gregory et al., 2004). Chylek andLohmann (2005) estimate that the warming overGreenland is 2.2 times the global warming (a result ofclimate change feedbacks near the poles), so that thecritical warming over Greenland could be reachedwith a global warming of only 1.2°C.

The rate at which Greenland ice could melt – andtherefore sea level could rise – is still an open ques-tion. The last IPCC report assumed a relatively sim-ple model with conservative estimates using the dif-ference between melting and snowfall, and con-cluded a duration for melting of several millennia(IPCC, 2001a). But that report did not consider thedynamic flow processes discussed above, which havesince been observed and could imply a much fasterreduction of the ice. This process is not taken fullyinto account in present ice models.

For the Antarctic ice masses the 2001 IPCC reportpredicted no melting, but, in contrast, a slight growthof ice due to increased amounts of snowfall. Newdata, however, also indicate a mass loss in the Antarc-tic and a dynamic response of the ice, especially in thesmaller West Antarctic ice sheet. In February 2002there was a spectacular collapse of the millennia-oldLarsen B ice shelf off the Antarctic peninsula afterwarming in this region (Fig. 3.1-3). This has no directeffect on sea level, because ice shelves float on thesea and their mass displaces a corresponding amountof water. But it evidently has effects on the continen-tal ice: the ice flows behind the Larsen B ice shelfwhich flow down from the continent have stronglyaccelerated since then (to up to eight times the speed:Rignot et al., 2004; Scambos et al., 2004).The floatingice shelves hang in part on projecting rocks, henceimpede the flow of the ice into the sea. The flow of

continental ice has also accelerated in other areas ofthe Antarctic, for example, in Pine Island Bay (Rig-not et al., 2002). In addition, it has been shown thatthe melting rate of the ice flow where it reaches thesea is very sensitive to the sea temperatures: per0.1°C rise in the water temperature the melting rateincreases by one metre per year (Rignot and Jacobs,2002). Thus, if the water temperatures around theAntarctic increase or if large ice shelves like the RossIce Shelf should one day disappear, then one has toassume that there will be a corresponding accelera-tion of the flow of the West Antarctic ice sheet.

Latest data from the GRACE satellite, which canprecisely measure anomalies in the gravitationalfield, indicate a shrinking of the Antarctic ice massesby 152km3 per year over recent years.This equates toa contribution to sea-level rise of 0.4mm per year(Velicogna and Wahr, 2006). The head of the BritishAntarctic Survey, Chris Rapley, has called theAntarctic in this respect an ‘awakened giant’.

Overall, the new observations suggest that the lastIPCC report could have underestimated the futuresea-level rise. A dynamic disintegration of the icesheets could possibly occur within a time frame ofcenturies instead of millennia. Unfortunately thepresently available ice models do not permit a reli-able prognosis for the further development of the icesheets.This uncertainty weighs even heavier because,with the positive feedback processes, the deteriora-tion of the ice sheets will be difficult to stop once ithas begun. These feedback processes include thelubrication of the undersides of glaciers with meltwa-ter from the surface and the frictional heat due tofaster flowing, as well as the lifting of shelf ice fromits resting points due to sea-level rise.

24

20

16

12

819841979 1989 1994 1999 2004

Year

Mel

t ar

eaA

pril

-Sep

tem

ber

[mill

ion

km2 ]Melt area

1992 2005

a b

FFiigguurree 33..11--22Extent of melt area on Greenland according to satellite data. The two extreme years 1992 (after the eruption of Pinatubo) and2005 are shown (a), and the development over time (b).Source: Steffen and Huff, 2005


3.1.1.3Further contributions to sea-level rise

Other contributing factors to global sea level are pri-marily the thermal expansion of water and the melt-ing of smaller mountain glaciers. Regional sea levelsare also influenced by changes in ocean currents andby geological processes (local uplift or subsidence ofland masses). As long as the global trend is small theregional processes can still predominate. Satelliteand water gauge measurements indicate that in spiteof global sea-level rise there are still regions withfalling sea level, e.g., in the Indian Ocean and aroundthe Maldives (Cazenave and Nerem, 2004). But ifglobal sea-level rise accelerates, it will eventuallyovercome the local effects and result in an overallrise.

According to water-gauge measurements, sealevel on the coasts has risen globally by 20cm since1870. That rise has accelerated throughout the 20thcentury, whereas the rate of rise was still near 0 at thebeginning of the 19th century (Church and White,2006). Over the past few millennia, according to geo-logical data, sea level hardly rose at all (Peltier, 2004)– this is also confirmed by analyses of water levels atthe time of the Roman Empire (Lambeck et al.,2004). Since 1993 it has been possible to measure sealevel globally and precisely from satellites – over thistime frame a rate of rise of 3cm per decade has beenrecorded (Fig. 3.1-4). Up to 5mm of the recent risecould be a fluctuation due to the eruption of thePinatubo volcano in 1991 (Church et al., 2005). Inde-pendent estimates of the individual contributions

currently give values of 1.6cm per decade (Willis etal., 2004) due to the warming of seawater, and 0.5cmper decade from mountain glaciers and smaller icemasses outside of Greenland and the Antarctic(Raper and Braithwaite, 2006).This leaves about onecentimetre per decade for the two large continentalice masses, which is consistent with the discussion inSection 3.1.1.2. In light of the uncertainties in theindividual contributions, however, it is still too earlyto derive a definitive balance of the present sea-levelrise.

The various scenarios of the 2001 IPCC reportyielded a rise of 9–88cm from 1990 to the year 2100.The lower of these values lie clearly below the rate ofrise already measured. This also suggests that theIPCC has so far underestimated sea-level rise.

3.1.1.4New estimates of sea-level rise

The physics of the observed dynamic processes in thecontinental ice discussed above are not adequatelyunderstood, and present continental-ice models donot yet consider these processes to a sufficientextent. There is an urgent need here for furtherresearch (Section 3.5). Improved estimates are diffi-cult given the present state of knowledge, and arepossible only with large uncertainties. Such an esti-mate, necessarily very rough, is attempted in the fol-lowing.

Sea-level rise up to the year 2300 is considered,with a stabilization of warming at 3°C above the pre-

36

FFiigguurree 33..11--33The Larsen B ice shelf off the Antarctic Peninsula in satellite photographs on 31 January (a) and 5 March 2002 (b).Source: NSIDC, 2002


industrial value. The comparatively long time rangewas chosen because of the intrinsic time scales of therelevant processes, amounting to several centuriesboth for the melting of ice sheets and for thermalexpansion of seawater. After stabilization of thegreenhouse gas concentrations and the climate onthe surface, the sea level will continue to rise for cen-turies. To estimate the impacts of anthropogenicemissions during the coming decades, therefore, aconsideration only to the year 2100 is not enough.

At a medium climate sensitivity of 3°C, this sce-nario corresponds to the effect of a doubling of thepreindustrial CO2 concentration, or a CO2 equivalentof 560ppm. If the worldwide contribution of CO2 tothe radiative forcing due to anthropogenic green-house gas emissions remains at 60 per cent, the560ppm CO2 equivalent would correspond to a sta-bilization at 450ppm of CO2.• Thermal expansion: For this the values of the

IPCC are adopted (0.4–0.9m: IPCC, 2001a, theirFig. 11.15a), which are derived from model simu-lations for a scenario with doubled CO2.

• Glaciers: For the volume of all glaciers outside ofGreenland and the Antarctic the same IPCCreport gives a sea-level equivalent of 0.5m; with3°C of global warming one could expect a loss of80 per cent of the glacial mass for the year 2300.Amore recent study (Raper and Braithwaite, 2006),however, uses half of this value; therefore a rangeof 0.2–0.4m is applied.

• Greenland: The model presented by IPCC (2001a)for Greenland with a local warming of 5.5°C(which is a plausible value with 3°C global warm-ing: Chylek and Lohmann, 2005) gives a sea-levelrise contribution of 0.9m by the year 2300. Thedynamic mechanisms mentioned above, however,

are not considered, so this value represents alower limit; therefore 0.9–1.8m is assumed here.

• Antarctic: The behaviour of the West Antarctic icesheet (WAIS) is critical for the Antarctic. In 2001,IPCC considered the decay of this ice sheet to bevery unlikely, because the then existing modelssuggested that the continental ice did not react tochanges in the ice shelves floating in the adjacentsea.This now has to be considered as disproved, asthe observations discussed above show.The disap-pearance of further ice shelves (like Larsen B: Fig.3.1-3) due to warming of seawater means that themelting of the WAIS must be feared with a similartime frame as Greenland. For this, 1–2m of sea-level rise is assumed by the year 2300. At a con-stant rate this corresponds to the disappearance ofthe WAIS in a time frame of 900–1800 years –some glaciologists consider that broad destructionis even possible within 300–400 years.

The net result is a rise of around 3–5m by the year2300. The value of 3m corresponds to a loss of one-sixth of each, the Greenland and the West Antarcticice sheets; 5m corresponds to one-third of each(Table 3.1-1).

FFiigguurree 33..11--44Global sea-level rise asrecorded by satellitemeasurements (upper linewith its linear trend), withthe projections of the IPCC(2001a) and its range ofuncertainty.Source: Cazenave andNerem, 2004

5

4

3

2

1

0

1995 2000 2005Year

Sea

leve

l ris

e [c

m]

Satellite measurementsTrend

IPCC Scenarios

Range of uncertainty forIPCC Scenarios

TTaabbllee 33..11--11Estimated global sea-level rise by the year 2300 with globalwarming limited to 3°C (explanation in text).Source: WBGU

Mechanism Rise in m

Thermal expansion 0.4–0.9

Mountain glaciers 0.2–0.4

Greenland 0.9–1.8

West Antarctica 1–2

Total 2.5–5.1


The question arises whether these numbers areconsistent with today’s observed sea-level rise rate of3cm per decade. Due to inertia and nonlinearity, andthe initial slow start-up of the rise, this cannot yet beanswered. At today’s measured rate of rise, therewould be an increase in sea level of only about 1m by2300. The present rise, however, is a response to only0.7°C global warming. At 3°C warming a pace fourtimes faster is plausible for the rate of rise and wouldbe consistent with the range estimated above.

This rough calculation, which does not represent aworst-case scenario, underscores the potential riskposed by sea-level rise, which could emerge to be oneof the most severe consequences of global warming.More precise and robust estimates are thereforeurgently needed. Research needs arise, above all, inthe areas of continental ice mass dynamics and thedynamics of the ocean (especially ocean mixing), inorder to reduce the uncertainty in the estimation ofthermal expansion (Section 3.5).

3.1.2Stronger tropical cyclones

Ocean-related impacts of climate change threatenhumankind and ecosystems not only through the risein sea level, but also through extreme weather eventssuch as tropical cyclones. The 2005 hurricane seasonbroke a series of records: not since the beginning ofrecord keeping in the year 1851 have there been somany tropical cyclones in the Atlantic (27, six morethan the previous record), have so many grown to fullstrength (15, four more than the previous record),and have there been three hurricanes of the mostdestructive category – category 5. A more intensivehurricane than Wilma, with a central pressure of only882mb on 19 October 2005 has never been measured.And with Vince, the first tropical storm to approachEurope was seen; it developed into a hurricane nearMadeira on 9 October 2005, and made landfall inSpain after weakening.

The hurricane season of 2004 was already extraor-dinary. For the first time Florida was hit by four hur-ricanes in one year, and for the first time Japan expe-rienced ten typhoons, as hurricanes in the Pacific arecalled. Of even greater interest for climatologists wasthe fact that in March 2004, for the first time, a hurri-cane developed in the South Atlantic: Catarina. Itformed in a region off the Brazilian coast, where asimulation calculated by the British Hadley Centrehad indeed previously predicted that hurricaneswould originate due to global warming (Met Office,2006).

The question arises whether there is a connectionbetween global warming and hurricanes. The central

statement regarding this in the last IPCC report wasthat an increase in the number of tropical cyclonesdue to global warming is not to be expected, and thatobservational data also show no significant trend inthe number of these storms.

Since this IPCC report was submitted there hasbeen a series of new studies on this topic.They do notexactly contradict the IPCC statement, but theythrow a completely new light on the question above,whereby the number of tropical storms is no longerthe focus of interest but their strength. The twoaspects are determined by different factors. Tropicalstorms arise from a small disturbance (such as a thun-derstorm) over the tropical ocean. In the Atlantic thisdisturbance often originates on the African conti-nent. What controls the frequency of such ‘embry-onic’ hurricanes is not yet fully understood, but thereis no evidence of a direct influence of global warmingon this process.

The further development of a tropical storm afterit has begun is, however, strongly determined by itssurroundings, i.e., by the sea temperatures and theatmospheric circulation.The sea temperatures in par-ticular are affected by anthropogenic warming.Whether the atmospheric circulation changesbecause of the warming, and to what extent this pro-motes or hinders the development of hurricanes isstill unclear. Here one is dependent on simulationswith global models, which, however, still have weak-nesses with respect to the resolution of hurricanes.The following points are well-supported by measure-ment data:1. Warmer sea temperatures lead to stronger hurri-

canes with more precipitation.2. Sea temperatures in the tropics during the rele-

vant season (around June to November) haveincreased and, in both the Atlantic and Pacific, areat their highest level since the beginning of mea-surements, which (although with decreasing qual-ity) extend back into the 19th century.

3. The energy of hurricanes has increased both in theAtlantic and Pacific, to their highest values sincethe beginning of reliable data in the 1950s. Whilethe total number of tropical storms has hardlychanged, the number of especially strong hurri-canes (category 4 and 5) has clearly increased.

The first point is well supported theoretically: warmtemperatures are an energy source for hurricanes,which is why they are a tropical phenomenon. Thisfact is routinely applied in the predictions of theNational Hurricane Center. Emanuel (2005) has ver-ified this connection based on measurement datasince 1950. He also defined an index for the strengthof a hurricane, the ‘Power Dissipation Index’ (PDI),which is the cube of the wind speed added over theextent and duration of a hurricane.An increase in the

38


PDI therefore is seen for stronger, larger or longer-lasting hurricanes. The PDI can be interpreted as anapproximate measure for the destructive potential ofa hurricane.

Figure 3.1-5 shows the increase of the PDI overrecent decades in the Atlantic; there is a similardevelopment in the Pacific. In addition to theincrease, the connection to the globally averagednear-surface air temperature is clearly recognizable.The increase of the PDI with temperature, however,is much stronger in the data than would follow fromthe theory of hurricane energy. This discrepancy isnot yet understood. A conceivable hypothesis is thatthe warm surface layer is thickened, so that the quan-tity of heat accessible for the hurricane increases outof proportion with the temperature (Scharroo et al.,2005).

Another study (Webster et al., 2005), using satel-lite data, has shown that the number of category 4and 5 hurricanes since 1970 has almost doubled glob-ally (that is, in the Pacific, Atlantic and IndianOceans), although the total number of tropicalstorms shows no significant trend during this time.This again confirms the statements of the IPCC(2001a), whereby the number does not change, and ofEmanuel (2005), whereby the strength increases.

In several studies a working group in Princetonhas investigated how global warming affects a hurri-cane model that is regularly employed for predic-tions by the National Hurricane Center (Knutsonand Tuleya, 2004). The model was run under bound-ary conditions from several global climate models,both for today’s climate and for a warming scenario.In these studies the frequency distribution of the hur-ricanes shifted clearly toward the stronger storms –the strongest hurricanes, those of category 5,occurred three times more often in the warming sce-

nario than in the reference climate. Because globalclimate models themselves so far do not have suffi-cient resolution to describe hurricanes very well,these studies, with a regional, high-resolution predic-tion model are the strongest tools available to datefor simulating the future development of thesestorms.

Theory, observational data and model calculationstherefore indicate that climate warming leads tostronger hurricanes. The effects revealed by mea-sured data are even stronger than theoreticallyexpected. With a warming of the tropical sea-surfacetemperature of only 0.5°C the hurricane energy hasincreased globally by 70 per cent in recent decades,and by even more in the Atlantic (Emanuel, 2005).Anew data analysis also confirms that the temperaturerise is the main reason for this observed energyincrease, while other factors play a minor role(Hoyos et al., 2006).

Yet, there are a few hurricane researchers in theUSA who attribute the extreme year of 2005 to a nat-ural cycle alone: to a fluctuation of the Atlantic cur-rents (‘thermohaline circulation’), which is discussedin Section 2.1.3. This is, so far, also the position of theNational Hurricane Center of the USA. These hurri-cane researchers, however, do not reject the connec-tion between higher temperatures and stronger hur-ricanes, rather they dispute that the warming itself isanthropogenic, and some of them even disputeanthropogenic climate change in general. Some stud-ies will appear in the near future that analyse theanthropogenic contribution to the increased Atlantictemperatures more accurately.

A natural cycle, in addition to global warming,could have in fact contributed to the extreme year of2005 in the Atlantic. But such a cycle cannot explainwhy the temperatures are higher now than ever since

0.6

0.4

0.2

0

-0.2

-0.4

2.0

2.5

1.5

1.0

0.5

0

-0.619201900 1940 1960 1980 2000

Year

Tem

per

atur

e d

evia

tion

[°C

]

Pow

er D

issi

pat

ion

Ind

ex (P

DI)

Tropical Atlantic

Global mean surfaceair temperature

PDI

FFiigguurree 33..11--55Temporal development ofthe energy of tropical storms(Power Dissipation Index –PDI, red) and the averagesea-surface temperature inthe tropical Atlantic fromAugust to October (blue).For comparison the patternof the globally averagednear-surface air temperatureis shown (dashed grey line).Source: after Emanuel, 2005


the beginning of measurements (and than the lastmaximum of the cycle in 1950) nor can it explain therise in the Pacific. There, where the majority of trop-ical storms occur, their energies have also shown anincreasing trend for decades. In addition, theobserved temperature development in the tropicalAtlantic lies within the range of the global warmingtrend (Figs. 3.1-5, 2.1-1, and 2.1-2), and is consistentwith that derived by modelling calculations as aresult of anthropogenic emissions.

To resume, it can be said that among hurricaneexperts (most of whom are specialists in weather pre-diction and not climate research) there is a consensusthat warmer sea temperatures strengthen tropicalstorms. Among climate experts there is a consensusthat anthropogenic warming has contributed signifi-cantly to observed warming in the tropical oceans. Acausal connection between global warming andstronger hurricanes is not proven by this and requiresfurther research, but it has to be considered as verylikely given the present state of knowledge.

3.2Impacts on coastal regions

The consequences of climate change, whether in theform of sea-level rise or through greater frequencyand force of extreme weather events, will directlyaffect the future development of coastal regions. Theworldwide length of coastlines (excluding small pro-trusions of less than a few kilometres) is on the orderof around one million kilometres. Coastal regions areof extreme importance for humankind. They offersettlement areas, are centres of economic activity(Turner et al., 1996) and, not least, harbour a richabundance of biological diversity.

The direct effects of climate change, such as theextent and rate of sea-level rise, presently cannot beprecisely determined. But it is very probable that thethreat to coastal regions will increase considerably, aswill the number of people affected by climatechange.This is an obvious result of the fact that largenumbers of settlements are located near the coasts.Eight of the world’s ten largest cities today lie on thecoast (UN, 2004), and according to estimates 21 percent of the world’s human population live less than30km from the sea (Cohen et al., 1997; Gommes etal., 1998). The great attraction of coastal regions isalso reflected in the large growth rates of populationsthere, which is around twice the global average(Bijlsma et al., 1996). The worldwide trend towardurbanization will amplify this development in thefuture. By the year 2030 approximately 50 per cent ofthe world population could be living within 100km ofthe coasts (Small and Nicholls, 2003).

How sea-level rise and weather extremes due toclimate change will affect coastal regions and soci-eties depends primarily on the kind and number ofaffected natural and social systems. The natural sys-tems will mainly be represented by river deltas, low-lying coastal plains, coral islands and atolls, barrierislands and lagoons, beaches, coastal wetlands andestuaries (IPCC, 2001b). The following sectionsexplore in detail which biogeophysical and socio-economic impacts can be expected and to whatextent people are threatened.

3.2.1Biogeophysical impacts

3.2.1.1Inundation due to sea-level rise

The rise in mean sea level will result in the inunda-tion of coastal areas and island groups in severalregions of the world. Inundation here is defined asthe permanent covering of land areas with water (asopposed to temporary, episodic flooding). Withoutcounter-measures the result will be the irretrievableloss of this land.

In order to be able to estimate the total extent ofthe regions endangered by sea-level rise, Brooks etal. (2006) have compiled data on the global land-sur-face distribution with respect to the elevation abovesea level. Figure 3.2-1 illustrates that large regions liewithin the range of one metre above high water.Above the one-metre line, the land-surface distribu-tion rises as an almost linear function of the elevationabove the mean high water line. At an elevation ofonly 20m above sea level a total land area of 8 mil-lion km2 would be affected.

40

Land

are

a [m

illio

n km

2 ]

10

8

6

4

2

050 10 15 20Elevation above MHWL [m]

FFiigguurree 33..22--11Distribution of land area, excluding Antarctica, as a functionof elevation above present mean high water (MHW).Source: ISciences, 2003

41Impacts on coastal regions 3.2

For purposes of illustrating the spatial distributionof these land areas, examples will be shown of regionsthat lie at elevations within 2m and within 20m abovesea level. A rise of 20m (Fig. 3.2-2) represents anextreme scenario that could result from anthro-pogenic warming over a time frame of around 1000years, in the event that the ice sheets of Greenlandand west Antarctica should melt for the most part(Section 3.1.1). This long time frame has to be con-sidered with sea-level rise because the relevantprocesses, such as melting of the ice sheets and mix-ing of the ocean, are slow geophysical processes.Because of the physical inertia in the marine systemthese processes will first come to a standstill cen-turies after stabilization of the greenhouse gas con-centrations and the surface climate.

The particularly threatened areas in Europe witha rise of 20m would be mainly eastern England, thePo Delta in northern Italy, and the coastal strips run-ning through Belgium, the Netherlands, north-west-ern Germany, and into northern Denmark (Fig. 3.2-2).

A sea-level rise of 2m (Figs. 3.2-3 and 3.2-4) couldoccur in the coming century. As an illustration of theeffects, Figure 3.2-3 depicts regions on the North Seaand the northern European coast. Because this kindof illustration is based on the absolute elevationabove sea level, it also includes areas that are pro-tected by dikes today. Some densely populated areasin the Netherlands, England, Germany and Italytoday already lie below the normal high-water level(EEA, 2005). For these regions the sea-level rise isespecially threatening. Here the question of the rateof change takes on a special importance, because amore rapid rise could hamper the implementation ofadaptive strategies (Brooks et al., 2006).

In Asia, with a sea-level rise of 2m (Fig. 3.2-4), forexample, the densely populated river delta of theGanges-Brahmaputra-Meghna with its network of230 rivers would be affected. The total river regioncovers an area of 175 million hectares and stretchesfrom India and Bangladesh to Nepal, China andBhutan (Mirza et al., 2003). Approximately 129 mil-lion people presently live in this river delta(Woodroffe et al., 2006), with a large portion of themin rural areas. With Dhaka and Kolkata (formerlyCalcutta) there are already two fast-growing megac-ities here, that is, cities with more than ten millioninhabitants.

3.2.1.2Flooding as a result of storm surges

In most cases the most destructive results of sea-levelrise will not be from the very slow rise of the mean

water level, but in the increasing occurrences ofstorm surges.

The origin of storm surges is often related to theinterplay of storm systems and tides. When stormspush water onto the coasts at high tide it can lead tothe flooding of large areas of land. Especially in riverestuaries damage can occur over large inland dis-tances (SwissRe, 1998). The word flooding heredescribes a temporally limited, partial or completewater cover of normally dry areas.This can be causedby the rise of surface water (still or flowing) over itsbanks, as well as by the results of strong precipitation(Münchener Rück, 1997).

Sea-level rise increases the exposure of coastalinhabitants to storm surges and storm waves, andwith it the risk of flooding. The destructive force ofthese kinds of weather extremes increases as a directconsequence of sea-level rise (Jimenez and Sanchez-Arcilla, 1997). Higher waves will more easily reachthe original coastline and also penetrate fartherinland. Even the water levels of the two-metre sce-nario exceed today’s standards for coastal defencestructures. Although Great Britain, for example, hasprotective structures that reduce the wave heightnear the coasts, it is questionable whether these mea-sures can provide long-term protection when theexceptional situation becomes the normal case. Ifwater depths should change or shores becomesteeper, which would result in a direct energyincrease for waves coming onto the land, then theexisting structures would no longer be sufficient ascoastal protection measures (Burgess and Townend,2004).

Additional factors could significantly increase therisks from flooding: changes in oceanic and atmos-pheric circulation patterns caused by climate changecan influence storms and their destructive potentialat regional and local scale. For example, an increasein the strength of tropical cyclones is anticipated(Section 3.1.2). Furthermore, climate warming couldcontribute to intensification of the hydrologicalcycle, which makes increases in the frequency andintensity of extreme precipitation events likely(IPCC, 2001a).

For the consequences of sea-level rise it is less crit-ical how much higher the average water height isthan how frequent certain high levels are reachedduring storm surges. This can be estimated by a com-parison of the expected average rise with statistics ofpast storm surges. Accordingly, the return periods,i.e., the time interval between certain critical gaugelevels, could be strongly reduced in the future (Loweet al., 2001). A model by the Hadley Centre for aregion in eastern England, based on a combination ofmeteorological data and an assumed sea-level rise of0.5m by 2100, shows a reduction in the return period

3 Sea-level rise, hurricanes and coastal threats42

FFiigguurree 33..22--22Coastal areas in Europe, parts of western Asia and North Africa. Areas below 20m elevation above the present mean sea levelare coloured red (not taking future coastal defence measures into account).Source: Brooks et al., 2006

FFiigguurree 33..22--33Coastal areas along the North Sea. Areas below 2m elevation above the present mean sea level are coloured red (not takingfuture coastal defence measures into account).Source: Brooks et al., 2006


of high-water events from 500 to 12 years (Lowe etal., 2001). Similar trends were calculated for thegreater New York City area, based on various climatescenarios. According to these, with a sea-level rise of24–95cm the return period of a 100-year flood in the2080s is shortened to 4 to 60 years (Gornitz et al.,2002; Section 3.3). When the return periods ofdestructive extreme events become too short, therepeated repair of damaged infrastructures would nolonger make sense, and they would have to be aban-doned.

Land-use changes such as the clearing of forests,urbanization and the conversion of alluvial plainsand wetlands can further increase the flooding risk,for example by weakening the water-retentioncapacity of the soil (Kundzewicz and Schellnhuber,2004). Straightened or built-up rivers without naturalforests and wetlands have less buffer capacity inextreme situations. The flow and sedimentationbehaviour of rivers influenced by engineering mea-sures often determines whether storm-caused flood-ing risks are amplified or attenuated.

3.2.1.3Coastal erosion

In contrast to floods, which are relatively rare eventswith sometimes catastrophic results, erosion repre-sents an episodically occurring process (Hall et al.,2002). During the erosion process, waves carry solidmaterials such as sand, mud and rocks away from thecoast and redeposit them for the most part in otherformations. A rise of sea level could accelerate theseerosion processes (Zhang et al., 2004; Stive, 2004). Inparticular with a small rise, erosion may prove to beof greater importance than flooding (Smith andLazo, 2001).

The erosion rates depend on the local conditions.If undercutting occurs along with the resulting col-lapse of steep coasts or coastal protection structures,erosion can represent a serious danger. In this con-nection it is important to note that above all the ratesof sea-level rise are relevant to changes in coastalmorphology. If sedimentation rates can keep up withthe sea-level rise, then a new equilibrium can beestablished and this can have a stabilizing effect onthe coastal processes. Sedimentation processes have

FFiigguurree 33..22--44Coastal areas along the Gulf of Bengal and in the Ganges-Brahmaputra-Meghna River Delta. Areas below 2m elevation abovethe present mean sea level are coloured red (not taking future coastal defence measures into account).Source: Brooks et al., 2006


contributed to coastal development since the begin-ning of the Holocene, and have been responsible forthe preservation of land areas, especially during theinundation of river deltas (Brooks et al., 2006). How-ever, if the sea-level rise accelerates so quickly that anew equilibrium cannot be established, or if sedi-mentation rates are significantly reduced due tomanagement measures, then a loss of coastal stripswill probably result. A well-known example is theNile, where the sedimentation rates were decreased,primarily by the construction of the Aswan Dam,which led to accelerated erosion of the northern NileDelta by tides (Stanley and Warne, 1998).

Many authors, including Zhang et al. (2004), referto the Bruun Rule (Bruun, 1962) in their prognosesof the erosion of coastal regions caused by sea-levelrise. This states that the erosion rates are approxi-mately 50–100 times higher than the relative rate ofsea-level rise, that is, a sea-level rise of 1m wouldresult in the loss of a 50–100m wide coastal strip.Opinions about the general applicability of theBruun Rule, however, vary widely, because it is basedon the assumption of a simple, two-dimensional sys-tem, and the establishment of a sedimentation equi-librium in the bank area. These preconditions, how-ever, can hardly be assumed in real situations. Ittherefore must be concluded that for estimating theresults of erosion along the coastlines, more complexmodels have to be applied that also incorporate, forexample, the sediment transport along the coasts andchanges in sedimentation equilibrium, as would bethe case with a sea-level rise.

3.2.1.4Impacts on groundwater

The rise of sea level can also cause the groundwaterlevel of a coastal region to rise. This is determined inpart by geographic factors (e.g., elevation above sealevel), and in part by geological factors (e.g., proper-ties of rock and soil layers).

Above all, a rise in the groundwater level causedby sea-level rise could impact on river deltas up to20–50km inland. This estimate is based primarily onthe observation that groundwater along the coastsflows above a dense, landward-moving saltwaterwedge. This balance between freshwater and saltwa-ter is physically controlled by the relationship of thedifferent water densities. With a rise of sea level,therefore, the overlying groundwater would also rise(Barlow, 2003). This can lead to a saturation of thesoil. It produces impacts not only on the freshwatersupply of a region, but also on agriculture (saliniza-tion risk), the stability of foundations, and the safety

and functioning of dewatering and other under-ground systems such as subways.

In addition, a rise in sea level can promote theinvasion of saltwater in coastal aquifers (seawaterintrusion). Model simulations by Sherif and Singh(1999) concluded that a rise of 0.5m in the Mediter-ranean Sea would result in the intrusion of saltwater9km into the coastal aquifers of the Nile Delta. Withthe same rise, however, in the Gulf of Bengal, only azone of 0.4km landward would be affected. Theseprocesses would result in an increased salinization ofthe groundwater and surface waters, with consider-able impacts on agriculture and the drinking-watersupply.The intrusion of saltwater into the groundwa-ter reservoirs of coastal zones can already beobserved worldwide today, e.g. in China and India(Shah et al., 2000). Through the increasing over-exploitation of freshwater resources in the denselypopulated coastal zones, this process can also be con-siderably amplified.

Although sea-level rise can cause the salinizationof river estuaries and near-coastal groundwaterreservoirs, this process is determined by a number ofother factors. The run-off behaviour of precipitationand its contribution to groundwater recharging alsocontrols seawater intrusion in coastal regions. Anincrease of freshwater runoff can counteract seawa-ter intrusion. Seawater intrusion would have long-lasting effects. Some of these aquifers would requirehundreds to thousands of years to reach a new har-monic equilibrium with the new sea level (Barlow,2003).

3.2.1.5Biological impacts

Besides temperature increase and acidification, theexpected sea-level rise is an important additionalstress factor for the often highly species-rich terres-trial coastal ecosystems or near-coast ecosystems.Two particularly relevant ecosystem types are coralreefs (Section 2.4) and mangrove forests, becausethey not only harbour great biological diversity, butat the same time play an important role in coastalprotection. This latter was illustrated by the tsunamicatastrophe in December 2004 in the Indian Ocean:on coasts with intact coral reefs and mangrove foreststhe flood wave was slowed considerably so that thedamage was less disastrous (Fernando and McCulley,2005; Dahdouh-Guebas et al., 2005; Danielsen et al.,2005).

How coral reefs will respond to sea-level rise canbe derived by reconstructions of the past or by modelsimulations. The adaptive capacity of corals in pre-historic times varied greatly (Montaggioni, 2005).

44


The average vertical growth rate of coral reefs sincethe last ice age is reported as at most 10mm per year(IPCC, 2001b). But because the growth rate of coralsis influenced by many factors (Section 2.4), andcorals in this century will also be impaired by warm-ing, acidification and other environmental factors, aprognosis for the adaptive capacity of these ecosys-tems cannot be made with respect to the rising waterlevel.

Around 8 per cent of the coastlines worldwidetoday are bordered by mangroves. More than half ofthe mangrove forests have already disappeared(WRI, 2001). The observed decline can be attributedin large part to changes in human uses of coastalzones. A study on the changes of mangrove belts inthe Amazon region (Cohen and Lara, 2003) showsthat sea-level rise can also have a local effect on thedistribution of mangroves. The rise of sea level in thefuture will force the near-coastal mangrove belts far-ther inland. The mangroves, however, will only beable to survive in areas where enough space is left forthem adjacent to the intensive human land use. Forthe preservation of this valuable ecosystem it istherefore urgently necessary to maintain protectedareas or create new ones that include a wide bufferzone on solid land. With the help of the HadCM3model, Nicholls (2004) evaluated the sensitivity ofcoastal regions to flooding under the different SRESscenarios (IPCC, 2000). In every case the sea-levelrise results in the loss of wetland areas. This studyalso shows, however, that the direct destruction ofwetlands by people could exceed the losses caused byclimate change.

Changes in the tidal ranges and high-water levelscaused by sea-level rise are an additional burden forcoastal ecosystems. The consequences includechanges in water depths, light and temperature, andcurrent speeds, and a shift in the freshwater-saltwaterdistribution. These can lead to physiological burdensfor some animal and plant species that could thenrequire a habitat change. Studies show that evenminor seawater intrusions into coastal seas lead tolarge disturbances in the structure and diversity ofzooplankton populations. Accordingly, small salinitychanges can result in a decline in the biodiversity ofcoastal ecosystems (Schallenberg et al., 2003). Thefunctioning and preservation of ecosystems aretherefore not only threatened by flooding because ofsea-level rise, but also by changes in the frequencyand strength of seawater intrusions.

The DIVA model (DINAS-COAST Consortium,2004) is a new interactive tool for integrated analysisof the results of sea-level rise. The model simulatesthe effects of local sea-level rise (including tectonicrises and falls) on the ecosystems and populations ofthe coastal regions of the world, and incorporates dif-

ferent adaptive strategies. It is based on the analysisof the worldwide coast lines in more than 10,000homogenous segments according to morphologicaland socio-economic aspects, a self-developed exten-sive worldwide database, and a series of coupledmodules. For a scenario with a mean sea-level rise of50cm by the year 2100 the model reports a loss ofmore than half of the freshwater wetlands in coastalregions, around 20 per cent of the coastal forests, anda quarter of the mangroves.

3.2.2Socio-economic impacts

3.2.2.1Impacts on people

The multiple effects of sea-level rise on the naturalenvironment will have a major impact on people andthe systems they depend on. It is likely that some ofthese effects will interact, intensifying each other,such as floods and erosion-related events. For inhab-itants of coastal regions, sea-level rise will be thebiggest challenge posed by global climate change(IPCC, 2001b).

The extent of climate-related hazard will alsodepend on the extent to which the ecosystems of theaffected coastal regions have been exposed to priordamage. Pre-existing environmental problems ofteninteract with the impacts of climate change. Forinstance, flood risk may be increased by changes inland use (deforestation, settlement, etc.) in hydrolog-ical catchment areas, or degradation of coastalecosystems (coral death caused by marine pollution,logging of mangrove forests for building materialsand to clear land for aquaculture installations, etc.).Moreover, it has been observed that in some citiesthe land mass is sinking below sea level. Contributingfactors in this case include both the physical pressureof buildings and infrastructure and intensive urbanmanagement practices, combined with groundwaterextraction, drainage and building activity. Nicholls etal. (1995) estimate that, at their most extreme, localrates of subsidence may be as much as 1m perdecade. A rise in sea level then makes the risk offlooding in these regions even greater.The fact that avariety of factors are superimposed on each other –disappearance of natural barriers, sinking of landmasses below sea level, and rising sea levels resultingfrom climate change – increases the risk to popula-tions (Nicholls, 2003).

Based on 1995 population figures, there are cur-rently 60 million people living below 1m elevationand 275 million below 5m elevation above mean sea


level. If estimates are adjusted to take into accountforecasts of population growth, the figures for theend of the 21st century rise to 130 million (below 1melevation) and 410 million (below 5m elevation;Nicholls et al., 2005). A more recent study by Brookset al. (2006) arrived at similar findings (Fig. 3.2-5).

How people in threatened areas will ultimatelydeal with the challenges of an accelerated sea-levelrise is a complex and dynamic process. Migrationaway from threatened areas will depend on the par-ticular situation in a given locality, and can rangefrom planned migration based on risk assessmentsand economic considerations to the sudden displace-ment of people fleeing floods, storms, or sudden ero-sion-related events. Due to the likely increase inextreme weather events, the incidence of sponta-neous migration following natural disasters willprobably exceed that of planned migration (Brookset al., 2006). This is especially likely to happen wherea radical change in the landscape occurs and the costsof protecting the affected population become dispro-portionately high. People in low-lying coastalregions, especially river deltas and small island states,are particularly at risk in this regard (Nicholls, 2003).Studies show, for example, that unless costly protec-tive measures are put in place, a sea-level rise ofaround 0.5m would put 1.5 million people at risk inthe Egyptian governerates of Alexandria and PortSaid (El-Raey et al., 1999). In the case of Europe,estimates suggest that 13 million people would be atrisk in the event of a sea-level rise of 1m (EEA,2005).

There is a whole range of model simulationsdesigned to obtain a more precise estimate of thenumber of people exposed to flood risks. Using theFUND model, Nicholls et al. (2006), for example,have simulated the consequences of disintegration ofthe West Antarctic ice sheet and the resulting sea-

level rise of 5m over a period of 100 to 1000 yearsbeginning in the year 2030.The impact of coastal pro-tection measures was evaluated using cost-benefitanalysis. In all scenarios, population displacementreaches a peak between 2030 and 2060. Based on the(extreme) assumption that the ice sheet will disinte-grate rapidly within 100 years, up to 350,000 personsper year will be forced to leave their homes over aperiod of ten years. This would give a total of 15 mil-lion people. However, these figures account for amere 2–3 per cent of the total number of people atrisk, because they are based on the assumption thatcoastal protection measures will be implemented ona large scale. In another study aimed at estimatingflood risks, Hall et al. (2005) arrive at the conclusionthat, under the A1 and A2-SRES scenarios, the num-ber of people at risk in Britain in the 2080s comparedto 2002 would double, rising from 0.9 million to 1.8million.

‘Sea-level refugees’Whether coastal dwellers who are forced to leavetheir homeland due to climate-related environmen-tal changes (‘sea-level refugees’) return home or set-tle further away from the coast will depend on awhole host of factors. On the one hand, the decisionwill be influenced by whether coastal protectionmeasures are put in place and how effective or reli-able these are. On the other hand, the positionadopted by local and regional government will alsoplay a role, for example by discouraging or even pro-hibiting return to evacuated areas (Brooks et al.,2006). Actual numbers of sea-level refugees will ulti-mately be determined by the interplay of these fac-tors and measures.

In any case, in the long term sea-level refugees willneed to be resettled elsewhere, and this poses newchallenges for policy. This is especially true in thecase of the inhabitants of some of the low-lying atollssuch as the Maldives, the Marshall Islands, Kiribati,Tuvalu or Tokelau. These island states, with a totalpopulation of more than 500,000 (CIA, 2005), lie amere 2m above sea level on average and are there-fore at risk of becoming uninhabitable or disappear-ing completely as a result of climate change. Theirinhabitants face a constantly increasing risk of salin-ization and drinking water shortages and higher riskof storms and floods even if the 1m guard rail (Sec-tion 3.3) is successfully adhered to (Barnett andAdger, 2003). These factors are already making theirimpact felt: the first relocations to higher-lying landtook place in December 2005 on the Pacific island ofVanuatu. In this particular case, decreasing intervalsbetween storm surges had made it necessary to relo-cate the village of Lateu. The United Nations Envi-ronment Programme regards this case as probably

46

1.0

0.8

0.6

0.4

0.2

050 10 15 20Elevation above MHWL [m]

Pop

ulat

ion

[100

0 m

illio

n]

FFiigguurree 33..22--55Population living below a certain elevation above mean highwater (MHW) in 1995.Source: Brooks et al., 2006


the first formally recorded resettlement measure ofits kind, resulting directly from the consequences ofclimate change (UNEP, 2005).

Official programmes are already in place to tacklethe problem of sea-level refugees. New Zealand hasreached agreement with the governments of Tuvalu,Fiji, Kiribati and Tonga on immigration regulationsfor their inhabitants under the ‘Pacific Access Cate-gory’. Each year, a certain number of refugees whosestatus is a direct result of the consequences of climatechange are granted a New Zealand residence permit.A whole set of conditions is attached to obtaining aresidence permit under these arrangements, how-ever, and older people and poor people are currentlyexcluded (Friends of the Earth, 2005). The right ofsea-level refugees to be granted refuge in other coun-tries needs to be enshrined in international law (Sec-tion 3.4.2.3).

Threats to human health In coastal areas, the primary threat to the lives andhealth of large numbers of people is posed by stormsand floods. Even today, a total of 75 million people incoastal regions are exposed to the risk of storm-induced floods.Assuming a moderate climate changescenario with a sea-level rise of 0.4m by the 2080s,this figure would rise to an estimated 200 million(IPCC, 2001b; Patz et al., 2005).

When assessing health-related consequences ofstorm tides and floods, a distinction can be drawnbetween immediate, medium-term and long-termimpacts. Immediate impacts refers to impacts arisingduring the event itself and which are due to theeffects of flooding. These include death and injurydue to drowning or collision with hard objects, and tohypothermia and cardiac arrest (WHO, 2002). In thiscontext, the World Health Organization (WHO) hascalculated that in the year 2030 the relative risk ofdeath due to flooding in the coastal areas of theEUR-B Region will be 6.3 times higher than in thebase years 1980–1999 (McMichael et al., 2004).Affected countries in the EUR-B Region includesome of the former Soviet republics, several Balkanstates, Turkey, Poland and the EU accession statesBulgaria and Romania.

The medium-term impacts of floods manifestthemselves most notably in an increase in infectiousdiseases resulting from ingestion of or contact withcontaminated water (e.g. cholera, hepatitis A, or lep-tospirosis), or respiratory infections due to over-crowded accommodation (IPCC, 2001b). The lack ofproperly functioning sanitary installations and publichealthcare provision makes these risks even greaterin poorer countries. Following the floods inBangladesh in 1988, for example, the most commondiseases were diarrhoea and respiratory infections,

while the most frequent cause of mortality for all agegroups under 45 years was acute watery diarrhoea(Siddique et al., 1991).

In the longer term, the consequences of sea-levelrise could influence the frequency and distribution ofdisease vectors. Inundation of coastal regions, forexample, affects the incidence of mosquito speciesthat breed in brackish water, e.g. the malaria vectorsAnopheles subpictus and A. sundaicus in Asia. Floodscould, however, also destroy the natural habitat ofsome pathogens, such as the EEE virus (easternequine encephalitis virus) found in freshwaterswamp areas along the US coastline (IPCC, 2001b).

In addition, the rise in sea level and the conse-quences of storm surges and floods pose a risk todrinking water supplies and food security. This is amatter of increasing salinization of freshwater reser-voirs, which not only affects drinking water supply,but can also adversely affect agricultural productivityin the vicinity of the coast. At the same time, floodscan also lead to considerable crop losses, as for exam-ple in the case of the 1998 floods in Bangladesh,where rice losses accounted for more than half oftotal agricultural losses and resulted in annual agri-cultural production falling to a mere 24 per cent ofthe expected total. Potential consequences includefood shortages and undernourishment (del Ninno etal., 2001; WHO, 2002).

As a result of the shock and the consequences ofsuch events, floods can also have long-term effects onthe psychological wellbeing of the people affected.Loss of family members and friends, social networks,property and employment can lead to post-traumaticstress syndrome. This manifests itself in feelings ofanxiety, depression, psychosocial disorders, andindeed can lead to an increase in suicide rates. It mustbe taken into account that psychological problems ofthis sort may not emerge until months or years afteran event of this sort (WHO, 2002).

According to a study by the World Health Organi-zation, more than 150,000 people are already dyingevery year due to the consequences of climatechange (WHO, 2002). The primary causes of mortal-ity in this context are increased incidence of diar-rhoea, malaria and undernourishment. WHO esti-mates suggest that additional health risks resultingfrom climate change will more than double world-wide by the year 2030 (McMichael et al., 2004).Theseestimates are based on forecasts of a sharp increasein the relative risk of floods, with smaller increases inmalaria, undernourishment and diarrhoea. Althoughsmaller relative changes in these phenomena areforecast, they have the potential to bring about dis-ease on a much bigger scale. Infectious diseases thusseem to present a greater risk to humanity than thedirect impact of sea-level rise. However, the models


used at present do not take account of potentialinteractions among the various health risks.

3.2.2.2Economic damage

Assessing the economic impact of climate change oncoastal areas also presents scientists with consider-able challenges.To be able to make any statement onthe overall costs of the impact of sea-level related cli-mate change requires detailed analysis that is alsohighly disaggregated in geographical terms so as toenable estimation of the expected damage. Suchdamage may take a wide variety of forms, rangingfrom damage to property to costs arising due to lossof human life or loss of biological diversity andecosystem services.Table 3.2-1 gives examples of sec-tors of society affected by sea-level rise and of thedamage and losses to be expected.

In order to assess potential physical damage andimpacts on people, it must be borne in mind that alarge number of megacities will be affected by a risein sea level. Of the 20 megacities throughout theworld, 15 are exposed to the sea (calculated on thebasis of data from Klein et al., 2002; UN, 2004).Theseinclude Tokyo, Mumbai and New York. Since devel-opment of megacities often entails exacerbation ofexisting local environmental problems, such as low-ering of the groundwater level, these areas often lackany natural buffering capacity to balance the conse-quences of a rise in sea level. In such cases, drinkingwater supplies could be jeopardized.This is an exam-ple of what is termed ‘critical infrastructure’(Bruneau et al., 2003; DRM, 2006), a category thatalso includes transport, telecommunications andenergy supply networks, and emergency, rescue andhealth services; it further includes the retail sector,public administration, banking and finance. Criticalinfrastructure refers to institutions that fulfil vitalneeds and guarantee public safety, uphold law andorder and ensure provision of basic public servicesand a functioning economy (Commission of theEuropean Communities, 2005a). Disruption or dam-age to this infrastructure can result in supply bottle-

necks and significantly impair public safety (BBK,2006), and may even have a destabilizing effect on awhole region. For example, a gradual rise in sea leveland extreme events resulting from it could interferewith the functioning of major ports and at times bringthem to a halt altogether, with a knock-on effect onregional trade and transport networks. Geophysicalchanges to coasts are thus also likely to have large-scale economic impacts on neighbouring and inlandregions (Brooks et al., 2006).

In addition to the costs arising from physical dam-age or disruption to production, there are also costsresulting from the loss of ecosystem services. Forexample, the negative impact of sea-level rise oncoastal ecosystems can adversely affect local fishingyields (Brooks et al., 2006). In many countries, espe-cially poorer countries, the security of people’s liveli-hoods often depends directly on the yield from theseecosystems. Any disturbance in the freshwater bal-ance, for example by seawater intrusion (Section3.2.1.4), can also affect agriculture. Increasinggroundwater salinization has already damaged com-mon agricultural land on the islands of Tuvalu(Friends of the Earth, 2005). As well as jeopardizingfood supply, this also brings about a decline in localeconomic activity.

The overall costs of climate change include on theone hand the damage caused by climate change inmonetary units, and on the other the costs of adapt-ing to climate change. Adaptation measures must beimplemented in accordance with the principle of eco-nomic efficiency so that the benefits of the measures(in the form of damage prevented) outweigh thecosts (for example costs associated with constructionand maintenance of sea walls). In other cases, it maybe more sensible in strict economic terms to foregoadaptation measures altogether and accept climatechange-induced damage.A cost-effective portfolio ofstrategies will also depend on environmental andsocio-economic conditions in a given region, andthese can change over time. For planning strategiesand decision-making, it is important that categoriesof costs and benefits associated are thoroughlyexplored and taken into account (Section 3.4.1.1).

48

Categories Damage or losses

Infrastructure Buildings, transport infrastructure (roads, rail networks, ports,airports), energy infrastructure, coastal protection structures

Economic sectors Fisheries, agriculture, forestry (timber in mangrove forests),tourism, transport

Human wellbeing Mortality, spread of diseases, migration/displacement of people,loss of landscapes and cultural assets

Ecosystems Services from coastal ecosystems, biological diversity, includingsome species-rich islands, disruption of the freshwater/saltwaterbalance

TTaabbllee 33..22--11Classification of damagecaused by a rise in sea level.Source: adapted fromFankhauser, 1995

49Guard rail: Sea-level rise 3.3

A great deal of data is required in order to assessthe overall costs of climate change worldwide.Detailed information is particularly needed for iden-tifying and assessing potential damage. Availableinformation, however, is often far from comprehen-sive and indeed may be quite rudimentary, especiallyin developing countries. As shown in Table 3.2-1,damage may take a variety of forms and also includesgoods that are not traded on the market and there-fore have no price. This applies particularly to loss ofecosystem services and biodiversity, which may bequantified in economic terms with the aid of surveysand economic methods of estimation. However, thisis an area that is fraught with uncertainty.

People will not simply put up with the effects ofclimate change. They will protect themselves fromdamage by putting measures in place to adapt. Eco-nomic analysis must therefore include exploringcost-effective combinations of strategies. To do this,models are needed that can simulate not only climatechange but also national economic developmentworldwide. Although some models of this sort arealready in use (Fankhauser, 1995; Yohe et al., 1999;Darwin and Tol, 2001), they are based on highly sim-plified assumptions, with the result that estimates ofglobal costs can only be calculated very roughly andare therefore of limited expressiveness. Data fromregional vulnerability analysis, however, enable moreaccurate estimation of the costs of climate changedue to sea-level rise, at least in smaller areas (e.g. Box3.4-2).

3.3Guard rail: Sea-level rise

3.3.1Recommended guard rail

WBGU recommends the following guard rail:absolute sea-level rise should not exceed 1m in thelong term (even over several centuries), and the rateof rise should remain below 5cm per decade at alltimes. For comparison: total anthropogenic sea-levelrise up to now has been 20cm; current rates arearound 3cm per decade (Section 3.1).

3.3.2Rationale

The recommended levels are based on WBGU’s esti-mation that a higher or more rapid rise in sea levelwould in all probability cause damage and losses tohumankind and nature that exceed tolerable levels.

As is generally the case with guard rails, this estima-tion contains a normative component and is notsolely based on scientific principles (Box 1-1), giventhat there continues to be considerable uncertaintysurrounding the actual consequences of sea-levelrise. WBGU hopes that this proposal will stimulatebroad debate within society on what is an acceptabledegree of sea-level rise and stimulate furtherresearch on its consequences.

As in the case of WBGU’s climate guard rail onthe increase in global temperature (a total of 2ºC andnot more than 0.2ºC per decade; Box 1-1), the conse-quences of sea-level rise depend both on the overallfigure and on the rate. Effects on structures that arenon-moveable in the long term, such as cities andworld cultural heritage sites, depend to a greaterextent on the absolute figure, while the rate of risetends to be more important for dynamic systems suchas ecosystems, beaches and some coral atolls, whichare able to adapt to some degree. Between the two –in other words between the overall figure and rate –there is a variable degree of trade-off, in the sensethat a higher absolute value may be tolerated if therate is slower, while the maximum rate is tolerable atbest for a very short time.

Absolute riseIn order to justify setting an absolute guard rail forsea-level rise that must be adhered to even in thelong term, one must consider the consequences of apossible very slow rise in sea level. Based on currentknowledge, in the view of WBGU a rise of more than1m would be intolerable, because severe conse-quences would be virtually unavoidable even with avery long period of adaptation. This applies, forexample, to a whole series of megacities in closeproximity to the coast, such as New York, Lagos orKinshasa.

New York City consists of several islands andpeninsulas and has around 1000km of coastline(Bloomfield et al., 1999). Figure 3.3-1 shows the areasof southern Manhattan that would be inundated inthe event of a ‘one-hundred year flood’ (water levels3m above normal levels) at today’s sea levels. In thiscase, massive damage could be expected to occur,with flooding of important infrastructure includingsome subway stations. Statistically, if there is a sea-level rise of 1m, this storm surge level would beattained not just once a century, but every four years.A ‘one-hundred year flood’ would then extend cor-respondingly further into the streets of Manhattan.

Similar storm problems are to be expected inother cities and in large river deltas (e.g. the YellowRiver, the Yangtze, the Ganges-Brahmaputra, theMississippi or the Nile). In developing countries,


poor population groups are often concentrated inthese endangered areas.

In its first report, the IPCC listed a whole series ofisland states that would face a considerable threatfrom sea-level rise. Many small island states wouldlose a significant proportion of their land if the sealevel rose by 1m (IPCC, 1990). Some of the islandsare at risk of becoming uninhabitable due to stormsurges resulting from a sea-level rise of this magni-tude. Affected islands include, for example, the Mal-dives, Kiribati, Tuvalu and the Marshall Islands, witha total population of 523,000 people.These problemsare exacerbated by the increase in tropical cycloneintensity (Section 3.1.2). Around another 380,000people living on the Caribbean islands of Anguilla,Cayman Islands, Turks and Caicos Islands and theisland state of the Bahamas would also be affected bythis. Although some of these islands have highground of up to 65m above sea level, with the rise insea level, storm floods would penetrate further andfurther inland. In many cases, virtually the whole ofthe island’s infrastructure (e.g. airports, roads) islocated directly on the coast.

If the sea level rises by more than 1m, there is anadditional risk that cultural heritage sites will be irre-trievably lost. Cultural goods from the past possess

‘outstanding universal value’ (UNESCO, 1972). Inview of this fact, in 1972 UNESCO adopted the Inter-national Convention for the Protection of the WorldCultural and Natural Heritage. Some 180 countriesare now signatories to this Convention.An importantcomponent of world heritage is its universality; itbelongs to all individuals and peoples of the world,irrespective of the territory in which it is located.

Great importance should therefore be given toprotecting these world heritage sites. A sea-level riseof more than 1m would pose a direct threat for exam-ple to the 12th century Shinto shrine of Itsukushimain Japan and the 8th century Shore Temple in Maha-balipuram in India. Both are important religious siteswhose special character derives from their coastallocation. To protect these sites from sea-level rise,one option might be to consider removing the monu-ments to another site. This would involve at leastsome loss, as the monuments are symbolically andhistorically rooted in their present environment.

A sea-level rise of 1m would also put, among otherplaces,Venice and St. Petersburg at considerable risk.In the storms of 1966, when flood levels peaked at 2mabove normal, large areas of Venice were submerged.Homes and businesses were destroyed as a result, butso too were valuable works of art (Nosengo, 2003). In

50

FFiigguurree 33..33--11Inundated areas (blue) in lower Manhattan (New York) in a statistically typical one-hundred year storm event based on thepresent sea level. A sea-level rise of 1m would result in storm tides of this height approximately every four years.Source: Rosenzweig and Solecki, 2001; data based on USGS, U.S. Army Corps of Engineers, Marquise McGraw, NASA GISS

51Guard rail: Sea-level rise 3.3

St. Petersburg too, storms could have devastatingconsequences. A researcher at the European Bankfor Reconstruction and Development (EBRD) sug-gests that a storm-induced rise in water levels of 2.5mwould inundate around 10 per cent of the city, whilea rise in excess of this level could affect up to one-third of the city (Walsh, 2003). As a result of thesedangers, extensive projects are currently under wayto build protective structures; in the case of St.Petersburg, international funding is also involved.

Many valuable coastal ecosystems would also bethreatened by a sea-level rise of this sort, for examplethe Kakadu National Park in Australia and the man-grove forests of the Sundarbans National Park inBangladesh and India (UNESCO, 2006).

Rate of riseThe rate of sea-level rise should not overstretch theadaptive capacity of human society or marine andcoastal ecosystems.

The adaptive capacity of ecosystems can be esti-mated using the example of coral reefs, mangroveforests and beaches.The last great rise in the sea leveloccurred at the end of the last Ice Age, over theperiod from 18,000 to 5,000 years before present.Since then, the rate of rise has always been less than20cm per hundred years, and usually well below this(Walbroeck et al., 2002; Peltier, 2004). In theHolocene era, after this last great sea-level rise cameto an end, coral reefs, beaches, mangrove forests andother ecosystems were able to become establishedagain along the newly formed coastline.

The maximum vertical growth of coral reefs is esti-mated at 10cm per decade (IPCC, 2001b). If the con-ditions are highly favourable, they could thereforepresumably keep pace with this rate of sea-level rise.Future growth rates will be markedly slower, how-ever, due to ocean acidification and warming andother environmental stresses (Section 2.4).

The adaptive capacity of mangrove forests andbeaches is highly dependent on sediment accretion.Sand beach loss already observed along many coast-lines is considered to be a consequence of sea-levelrise (Leatherman, 2001). Ellison and Stoddart (1991)analyse the development of mangrove forests duringthe Holocene and arrive at the conclusion that in asituation where there is little sediment accretion,even the current rate of sea-level rise places exces-sive demands on adaptive capacity and will result inthe loss of mangrove forests. Other authors(Snedaker et al., 1994), meanwhile, argue that if thehabitat is favourable, retreat of mangrove forests fur-ther inland could enable them to accommodate aneven higher rate of sea-level rise. In many cases, how-ever, such favourable conditions will not be present.Based on a scenario with an almost linear sea-level

rise of 5cm per decade, the global DIVA model (Sec-tion 3.2.1.5) projects a continuous loss of mangroveforests whose adaptive capacity has thus alreadybeen exceeded. By 2100, according to this projection,a quarter of all mangrove forests would disappear.

According to the scenarios postulated by IPCC(2001a) the rate of sea-level rise towards the end ofthis century will be 3–7cm per decade, with up to13cm per decade in the worst-case scenario. In viewof these facts, WBGU recommends setting the guardrail for maximum sea-level rise at no more than 5cmper decade. It must be borne in mind, however, thateven compliance with this guard rail will not provideprotection from damage that is already significant, asis also the case with other WBGU guard rails (Box 1-1).

3.3.3Feasibility

The current and foreseeable rise in sea level is almostentirely anthropogenic, and hence its future develop-ment can also be influenced by humankind.The abil-ity to control it is limited on the one hand by the longtime-scale required (centuries) for a response interms of sea-level change. It is also limited by fore-casting difficulties and by the potential for stronglynon-linear behaviour on the part of the great conti-nental ice sheets. Nevertheless, the recommendedguard rails can be implemented, according to currentknowledge, by means of an appropriate climatechange mitigation strategy.

Stabilizing the global temperature at 2ºC abovepre-industrial levels, according to the mathematicalmodels, would result in a sea-level rise of around50cm in the long term (after 1000 years) simply dueto thermal expansion. Mountain glaciers would addapproximately another 20cm to this (Section 3.1.1.4).Prevention of large-scale melting of the continentalice sheets in Greenland and Antarctica would there-fore be critical for compliance with the guard rail.Further research must establish the limit that needsto be set as regards the rise in global mean tempera-ture in order to achieve this. It is conceivable that, inthe long term, it may be necessary to reduce the tem-perature to below the 2°C threshold again.

In this century, the guard rail for the rate of sea-level rise would only be breached by the more pes-simistic half of the IPCC scenarios (2001a); the moreoptimistic scenarios comply with the guard rail evenwithout climate protection measures. It should beborne in mind, however, that the currently observedrate of rise of 3cm per decade is already clearlyhigher than all of these scenarios (Fig. 3.1-4). It musttherefore be assumed that, in all likelihood, the IPCC


(2001a) has underestimated sea-level rise, and thatclimate mitigation measures are indeed required tocomply with this guard rail. Based on the assumptionthat the change in sea-level rise will be relativelysmooth and gradual, as depicted in all the scenarios,compliance with the guard rail for the rate of sea-level rise would mean a maximum rise in sea level ofaround 40cm in the 21st century. This would be dou-ble the sea-level rise that has taken place to date as aresult of human activity.

The climate and sea-level rise guard rails areclosely interlinked, since sea-level rise is directlycaused by global warming. In the next few decades,the climate protection strategies required to meet the2°C goal and to comply with the guard rails relatingto sea-level rise will most likely be similar and com-patible. Despite the long-term nature of sea-level riseand the uncertainties surrounding the behaviour ofthe continental ice sheets, these guard rails are notredundant. Even if the global warming guard rail isobeyed and lasting climate warming of 2°C takesplace, this would be enough to cause melting of theGreenland ice sheets, thereby breaching the guardrail on sea-level rise. For this reason, it is conceivablethat the guard rail on sea-level rise will lead to theimposition of strict limits on emissions, especially inthe long term, in other words in the coming centuries,in order to stabilize the continental ice sheets.

This is why it is vital, as regards emissions, toembark on a path that will lead to stabilization of theglobal temperature at a low level after 2100, and ifpossible well below 2ºC above the pre-industriallevel.The guard rail on sea-level rise therefore deter-mines in particular longer-term climate protectiongoals from the second half of the century onwards. Inthe coming decades, it is a key additional justificationin support of the 2°C goal. If, on the other hand, thecontinental ice sheets of Greenland and Antarcticawere to shrink suddenly and unexpectedly, the guardrail on sea-level rise could require tougher climateprotection measures than the 2°C climate guard raileven sooner. It thus gives particular grounds forcloser observation of the ice sheets in order to iden-tify dangerous developments in time.

3.4Recommendations for action: Develop andimplement adaptation strategies

In its previous reports on climate policy, WBGU hasmade it clear that priority should be given to strate-gies for preventing greenhouse gas emissions. How-ever, even if there is substantial success in preventinggreenhouse gas emissions and complying with theguard rail on sea-level rise, it will no longer be possi-

ble to prevent some of the effects of climate changeon coastal areas. Appropriate adaptation measuresare required in order to cope with these effects. Asregards strategies for adapting to sea-level rise andextreme weather events, WBGU focuses on twoquestions in particular:1. How can the anticipated destruction of coastal

infrastructure and settlements be coped with?2. How can provision be made under international

law to deal with land loss?

3.4.1Adapting coastal regions to the consequences ofclimate change

3.4.1.1Adaptation options: Classification and assessment

The extent to which the consequences of climatechange will give rise to damage in coastal areas andturn hazards into disasters varies considerably fromone region to another and depends on the vulnera-bility of the areas affected. This in turn depends onthe susceptibility and resilience of natural, social,infrastructural, economic, institutional and culturalsubsystems (Titus et al., 1991; Klein et al., 1999).Resilience in this context means the ability of subsys-tems to cope with repeated disruption so that keystructures and processes remain intact (Burton andLim, 2001; Burton et al., 2002; Adger et al., 2005).

Industrial countries will be better able to deal withhazards than developing countries, because theyhave more extensive capacities at their disposal, suchas an efficient institutional infrastructure, technicalknow-how and financial resources. HurricaneAndrew, for example, a category 5 event on the Saf-fir-Simpson hurricane scale, cost the lives of 23 peo-ple in the USA in 1992. A typhoon of comparablestrength that hit Bangladesh in 1991, meanwhile, ledto extensive flooding that resulted in 100,000 deathsand millions of refugees (Adger et al., 2005).

The large number of influencing factors and inter-actions makes it essential to develop adaptationstrategies that are tailored to the given context.Adaptation in this context needs to fulfil two pur-poses: on the one hand to reduce damage, and on theother to increase resilience of the above subsystems.There are basically three different options for adap-tation in response to the hazards outlined above:‘protection’, ‘managed retreat’ and ‘accommodation’(IPCC, 2001b).

52

53Recommendations for action: Develop and implement adaptation strategies 3.4

ProtectionProtection involves protecting coasts from rising sealevels by means of structural measures. These mightinclude ‘hard’ engineering measures such as con-struction of sea walls, dykes, or flood defence sys-tems, and ‘soft’ measures such as conservation orintroduction of protective coastal ecosystems (e.g.wetlands, mangroves, islands) or beach nourishmentas natural barriers. Hard structural adaptation mea-sures are exceedingly cost-intensive in terms of con-struction and maintenance. In addition, they increasestress on neighbouring ecosystems; for example, theyincrease the threat of wetland loss.Without interven-tion, wetlands will tend, as a rule, to migrate inland inthe event of floods. This autonomous adaptation isimpeded by sea wall construction, because areas onthe seaward side of sea walls become inundated,while on the landward side new wetlands are pre-vented from forming. In the case of US coasts, it isestimated that 50 per cent of all wetlands have dis-appeared as a result of this process (Titus, 1990). Inthe coastal regions of the EU, moreover, it has beenobserved that adaptation involving hard engineeringmeasures can trigger or accelerate erosion processesin neighbouring coastal areas. This in turn can signif-icantly impair the functioning of hard protectionmeasures (Commission of the European Communi-ties, 2005b; Brooks et al., 2006). Due to the multitudeof problems associated with hard structural adapta-tion measures, preference is given nowadays to ‘soft’measures wherever possible. Soft strategies interfereless with coastal ecosystems and permit a more flexi-ble response to sea-level rise, the extent of which isfraught with uncertainty. Ultimately, however, theeffectiveness of soft and hard measures will dependon the environmental and societal context.

Managed retreat Managed retreat means that use of areas in proxim-ity to the coast is reduced, or certain areas are relin-quished completely. In this context, strategies mightinclude moving buildings and settlements, and intro-ducing government regulation on the use of vulnera-ble areas. Retreat may be enforced by means of pub-lic order legislation, e.g. by regulating land use undernational construction and planning law. Anothermeans is to provide incentives in favour of the deci-sion to retreat voluntarily. Measures of this sortencourage households and private businesses to takeaccount of all costs relating to use of the coast in theireconomic decisions to invest.A targeted informationpolicy implemented by local public bodies could helpto enhance awareness of the implications of climate-induced risks.

In certain cases, a sensible option may be toactively support resettlement of people from thecoast to less threatened areas, for instance by provid-ing grants via the regional administrative bodies, orwithin the framework of international developmentcooperation.

The issue of resettling communities and their resi-dents arises in a very concrete way in the aftermathof a natural disaster, in other words, when infrastruc-ture has been destroyed over a large area.A decisionmust then be made as to whether reconstruction iseconomically sensible according to the prognosesrelating to future sea-level rise and the incidence andintensity of extreme weather events. The more resi-dents can rely on the government to share the costsof protection measures, the more they will beinclined to remain in threatened regions. If, however,each individual is confronted with the costs of pro-tection, the attractiveness of reconstructiondecreases and more people will opt to migrate to lessthreatened areas. In order to provide the right incen-tives in such instances, therefore, government (andinternational) aid for reconstruction must be tied toa corresponding relocation condition. Municipalities,too, must weigh up their adaptation options anddecide between protection and retreat. After a nat-ural disaster, they will tend to rebuild destroyedinfrastructure rapidly to be able to ensure continuityin public services. This is why it is important todevelop strategies for resettlement for threatenedareas before a natural disaster strikes (Brooks et al.,2006).

It is conceivable that, despite government incen-tives to encourage migration away from threatenedcoastal areas and an adequate information policy onthe part of public institutions, some people will notagree to relocate on a voluntary basis. In such a situ-ation, the government must decide whether it willpermit the affected people to remain and face dam-age to property and life at their own risk, or whetherit will forcibly relocate sections of the population.The latter option, of course, carries considerablepotential for conflict (Box 3.4-1).

Government measures encouraging migrationaway from coastal areas should go hand in hand withmeasures limiting migration of people into theseareas. For example, levying a tax that reflects thesocial costs resulting from migration of people intocoastal areas ensures that these costs are included inan individual’s calculation of the costs associatedwith migrating into the area, and thus become rele-vant for his decision.

Government regulation can thus fundamentallysupport the relocation of people in the desired direc-tion. It is nevertheless possible to provide false incen-


tives, for example, via interventions in insurance mar-kets. Due to the increasing incidence of floods andcyclones, economic adjustments are in fact to beexpected in the insurance markets: insurance premi-ums for flood damage will rise, and some privateinsurers will withdraw from the market. As a result,coastal areas lose their attractiveness as areas for set-tlement. If insurance premiums are kept artificiallylow by government subsidies, however, as is the casein the USA, prices are distorted and incentives tomigrate away from coastal areas are reduced.

AccommodationThe third strategy, termed accommodation, involvesmodifying land use and subsystems to ensure thatthey take account of the new threats. Residents ofthreatened regions continue to use the threatenedland, but without trying to protect it from inundation.This can take place, for example, by instituting disas-ter management systems (constructing emergencyrefuges, formulating plans of action, undertaking tar-geted public education and communication work). Itis likewise possible to modify land use, for exampleby cultivating varieties of grain that are resistant toincreasing salinization and inundation of the soil orby converting arable land to fish farming facilities. Inaddition, accommodation also includes engineeringmeasures (such as increasing the height of buildings,making cellars and buildings water-tight).

Portfolio approachIt frequently happens that these options are imple-mented as a combined set of strategies rather than asalternatives. A ‘portfolio approach’ is pursued inorder to respond adequately to the given conditionsin a particular region. One possible combination ofstrategies involves partial retreat, where protectivemeasures are applied only to areas where there is ahigh concentration of people, assets and functions.Flooding is allowed to take place in the other areas.Using this approach, implementation of protection

measures would be prioritized, focusing on politicaland economic hubs such as cities, towns and indus-trial areas.A particular focus of attention in this con-text is protecting the ‘critical infrastructure’, in otherwords, infrastructure so essential that its destructionhas a destabilizing impact on a country’s public lifeand economy.

Another strategy that might be considered is tocombine protection with accommodation. This couldinvolve, for example, aiming to increase coastalresilience through conservation of mangrove forestsas natural barriers. In the context of local land useplanning, setback areas could be created or extendedto permit ecosystems to shift landward, therebyenhancing an area’s capacity for autonomous adap-tation (Nicholls, 2003).

3.4.1.2Choosing adaptation strategies

Cost-benefit analysis may be used (Box 3.4-2) to helpchoose appropriate adaptation strategies for a spe-cific region. This requires comprehensive informa-tion on the state of coastal areas and on the impact ofhuman activities. Assessment of the interaction ofland and sea for commerce and industry, for portfacilities, buildings, groundwater and extraction ofconstruction material is also called for in this context(Kullenberg, 2001; SEEDS, 2005). The data requiredfor this purpose are gathered and evaluated in theframework of vulnerability studies (Burton andDore, 2000).

In contrast to prevention strategies, the effects ofadaptation projects are essentially local; in otherwords, they have no direct global environmental ben-efit. Moreover, because the extent of the environ-mental effects is fraught with uncertainty, ‘no-regret’measures should initially be identified and put inplace.These are measures that bring a net benefit forstakeholders irrespective of any actually occuring cli-

54

BBooxx 33..44--11

PPootteennttiiaall ffoorr ccoonnfflliicctt oovveerr rreesseettttlleemmeenntt

Depending on regional scenarios of threat, policy-makersmust consider the option of planned resettlement of popu-lation groups. However, a large number of projects in agreat variety of socio-economic and political contextsdemonstrate the many problems that can arise as a result ofsuch measures.As examples, dam projects, mining and infra-structure projects (e.g. the Three Gorges Dam in China, lig-nite open-cast mining in Garzweiler, Germany, road con-struction in metropolitan Manila, etc.) can be cited.

Although resettlement of endangered coastal residentsis generally a necessity in order to protect the affected peo-ple, there is also considerable potential for conflict in thissituation. For example, decisions to protect important infra-structure installations may amount to unequal treatment ofdifferent population groups (populations in the proximityof a protected installation will be protected along with it,while other settlements are evacuated). In addition, exacer-bation of conflicts relating to land use in the target area forresettlement may occur (conflicts between long-standingresidents and new settlers). Massive resistance is most likelyto occur, however, in regions where resettlement pro-grammes were used in the past as a repressive measure bythe government.


mate-induced losses. Such measures tend to wingreater support among affected stakeholder groupsbecause they take into account the uncertainties ofclimate change and yield desirable results even if cli-mate change were not to occur.An example might bea coastal region with pre-existing damage and a highpopulation density, for which a rise in sea level wouldexacerbate existing problems. Improving planningrelating to use of coastal areas would be an appropri-ate strategy here for dealing with sea-level rise.Moreover, it would still bring a positive net benefiteven if the anticipated effects of climate changefailed to materialize.

In practice, transaction costs, institutional failureor lack of information have frequently led to theshelving of such projects. Adaptation projects canhelp to dismantle these obstacles (Fankhauser, 1998).Implementation of integrated coastal zone manage-ment, for example, could help to improve theexchange of information among the different policy-makers and thereby make it easier to carry out pro-jects.

3.4.1.3Implementing adaptation strategies

Adaptation requires more than simply implementingengineering options. Not only is the choice of strate-gies influenced by a multitude of factors; the strate-gies themselves impact on the subsystems of theregion in which they are implemented. It is also nec-essary to reconcile the various responsibilities andinterests of participating or affected groups in society(Nicholls, 2003).

Risk managementRisk management provides an ideal means of imple-menting adaptation strategies. Risk managementplans designate persons responsible (public and pri-vate, at municipal, national and international level)for all stages of an event – before, during and after.They describe what measures should be taken(strategic versus tactical measures) at what point intime, and the manner in which the responsible per-sons should respond and to whom they should report(Boyd et al., 2005). In many instances, policy-makersdo not treat the issue of climate change as a priority,and as a result, climate-induced changes in the risksituation are not adequately taken into account.Risks are thus often graded as low and threats areconsidered as rather unlikely, with the result that thedesign of available risk management plans is inap-propriate. The example of Hurricane Katrina, whichcaused destruction on an unprecedented scale on theUS coast in August 2005, shows that inadequate plan-

ning can heighten the socio-economic impact of suchevents.

Formulating an appropriate risk managementplan should ideally be a cyclic process. In advance ofan extreme event, a planning phase (Phase 1) – inwhich preventative and reactive measures aredevised – is followed by a preparatory phase (Phase2). Measures included in this second phase are aimedat reducing the likelihood of potential hazards result-ing in disasters. This may be done by establishingaction plans, providing emergency training and con-ducting targeted public information and educationcampaigns, or by establishing agreements for inter-national cooperation in the area of disaster assis-tance and for dealing with environmental refugees. Ifan event actually occurs, the response phase becomesrelevant (Phase 3). This phase involves measures tobe carried out during and after an event. Such mea-sures include emergency response, measures to pre-vent consequential damage such as outbreaks of epi-demics, or implementing measures aimed at speedingthe recovery of affected areas. The reconstructionphase (Phase 4) concludes the process of managingsuch an event.All activities in this phase are aimed atrestoring normal system functioning, via disbursal ofinsurance payments, setting up temporary emergencyshelters or reconstructing the physical infrastructure.In addition to these four phases, problems are identi-fied with regard to the handling of the event and mis-takes are analysed. The experience gathered is thenevaluated in a new planning phase and implementedin the form of improved strategies (Boyd et al., 2005).

In the case of slow onset hazards, on the otherhand, the priority of risk management lies in regu-larly assessing the potential risks and identifying themost vulnerable individuals and regions. Risk man-agement involves adaptation to constantly changingconditions. A high degree of flexibility in terms ofstrategies is needed to achieve this. Such strategiesinclude in particular scientific monitoring, publiceducation and communication and legislative provi-sions (Boyd et al., 2005).

Integrated coastal zone managementIn order to take account of the highly complex, inter-connected nature of impacts, adaptation measuresshould be very broad, in other words, they should beenshrined in all key areas of policy. Examples arecoastal protection plans and strategies for sustain-able development.Another term used in this contextis ‘integrated coastal zone management’. As part ofthis system of management, data on both ecosystemsand social systems are collected and processed. Inte-grated coastal zone management as an instrumentfor managing risk in this context refers to a dynamicprocess developed and implemented on the basis of a

3 Sea-level rise, hurricanes and coastal threats56

BBooxx 33..44--22

CCooaassttaall mmaannaaggeemmeenntt oonn tthhee GGeerrmmaann NNoorrtthh SSeeaaccooaasstt

Global forecasts on the impact of sea-level rise are notdirectly applicable to regional or local circumstances. Evenwithin Germany’s coastal areas, major differences can beobserved as regards the hazard situation and socio-eco-nomic resilience. To be able to put expedient measures inplace to adapt to future consequences of climate change,therefore, small-scale, scenario-based studies need to beconducted that analyse both the natural and social charac-teristics of a given area. Studies of this sort have alreadybeen carried out for two regions of the German North Seacoast: for the island of Sylt and for the north-west Germancoastal region.

As a result of its particularly risky situation and eco-nomic productivity, the North Sea island of Sylt wasanalysed in the context of a study commissioned by the Fed-eral Ministry for Education and Research (BMBF) entitled‘Climate change: consequences for people and coasts’. Theisland is an open system with a negative sediment balance inwhich erosion processes occur that result in a continuousreduction in land mass. Sea-level rise is expected to acceler-ate these processes. The economic structure of Sylt is heav-ily biased towards tourism, which is concentrated on thewestern side of the island.

Various scenarios were devised in order to estimate theconsequences of climate change up to the year 2050. Thescenario presented here is based on an assumed local sea-level rise of up to 25cm and changes in wind conditions, tidalrange and swell (i.e. wave height, direction of approach andperiod). Extreme weather and its impacts on the naturalenvironment and socio-economic structures were not con-sidered, so further research is needed to explore theseaspects. Based on the results of the model simulations,changes in sediment transport are to be expected on thewest coast of the island, which would have a negative effecton the wave-attenuating impact of an offshore reef. Thethree municipalities of Rantum, Hörnum and Wenningstedtare likely to be worst affected by this.

The study recommends adaptation of these coastal areasby adopting a ‘portfolio approach’, in other words, by a com-bination of different component strategies. Consideration isgiven to the three components: protection, managed retreatand accommodation. A combination of ‘soft’ and ‘hard’coastal protection measures was identified as the optimumstrategy for protecting Sylt’s current coastal form.The mea-sures focus particularly on the environmentally sustainableoption of beach nourishment, which is currently alreadybeing implemented on the west coast of the island.

To evaluate the recommended adaptation measures ineconomic terms, cost-benefit analysis was carried out forthe west coast of the island.According to this analysis, adap-tation costs would consist primarily of the costs of addi-tional beach nourishment measures. At current values, thecosts involved in the period up to 2050 are estimated at €33million. The benefits of coastal protection in terms of pre-vented losses of assets, infrastructure, beaches and dunesmust be balanced against this.The current value of this ben-efit is estimated at €381 million. The results of the analysisfor this period thus show that the benefit-cost ratio ofcoastal protection for the island of Sylt is clearly positive.The scenario examined illustrates that the island of Sylt canprobably be protected effectively against a small rise in sealevel of 25cm by beach nourishment measures.

At the same time, it must be noted that Sylt represents aspecial case, with its particular geographical characteristicsand its very high concentration of assets resulting fromtourism. The recommended adaptation measures areunlikely to be applicable to the majority of the world’s othercoastal regions. Beach nourishment, for example, is onlypossible, and only makes economic and environmentalsense, if sand is available within the coastal region in suffi-cient quantities. In addition, the estimated adaptation costsare based on a coastal protection strategy aimed primarilyat protecting the main part of the island, which is higher-lying compared to the shore, from erosion. Beach nourish-ment would be unlikely to provide adequate protection forflatter sections of coast in the face of rising sea levels – espe-cially in the event of extreme weather situations.

Building on the experiences of the Sylt study, the project‘Climate change and preventive risk and coastal protectionmanagement on the German North Sea coast’ (KRIM) notonly examines the consequences of accelerated sea-levelrise for various stretches of coastline, but also the concomi-tant risks of extreme weather events. Using the same timehorizon of 2050, this study analyses future consequences ofclimate change together with possible social adaptationmeasures and their impact.

The KRIM project assumes a regional increase in tem-perature of 2.8°C, a local sea-level rise of 55cm, and changesin mean tidal range, precipitation, swell, and winter windvelocity and direction. In addition, extreme weather condi-tions with flood levels of +200cm are also taken intoaccount.

In order to analyse the consequences of this climate sce-nario for Germany’s north-western coastal region, theresulting risks of extreme weather events were calculatedand the costs and knock-on effects for the regional econ-omy of possible coastal protection strategies were setagainst these. Alternative strategies were compared usingcost-benefit analysis techniques. When assessing potentialdamage from storm surges, the KRIM project not only con-sidered environmental damage and economic losses fromdamage to assets and infrastructure, but also the resultinglosses to the national economy in terms of value-addedactivities, income and jobs. A mesoscale modelling tech-nique was used to assess the value of economic losses; inother words aggregated data based on official regional sta-tistics were used. Subsequent to the value assessment, losseswere calculated as a function of flood levels.

Among other things, the study examined dyke raisingand construction of a second line of dykes as elements of aprotection and accommodation strategy for the study areaof Wangerland. The location of Wangerland (a regionbetween river Weser and river Jade at the North Sea) is geo-graphically very exposed, with long coastlines requiringprotection, but it has relatively few assets. Based on the year2010 as the assumed investment year (base scenario), thecosts of dyke raising were estimated at €10.5 million (for anaverage increase in height of 0.75m over a 28km stretch ofdyke), while the costs of constructing a second line of dykes(variant II) were estimated at €20 million (for a 17kmstretch of dyke with a height of 3m above mean sea level).The current value of economic losses in the KRIM climatechange scenario, meanwhile, is estimated at €63 million(2000) – calculated using the flood simulations and for theperiod up to 2050. According to these calculations, the ben-efit-cost ratio is highest for the dyke-raising option, with theresult that action was recommended to implement thiscoastal protection measure.


coordinated strategy with the aim of managing envi-ronmental, socio-cultural and institutional resourcesso as to ensure sustainable conservation of coastalareas and ensure that they can be used in a variety ofways in future (Fankhauser, 1998; Yeung, 2001).

Coordinating the sectoral, competing and in somecases overlapping competences of the various deci-sion-making tiers and specialist areas within theadministration presents a major challenge as regardsdevising an integrated coastal management system.Institutional fragmentation often gets in the way ofproviding adequate responses. WBGU therefore re-commends creating integrated institutions that bringtogether all the key competences. Such institutionswould also facilitate reconciliation of the diverseinterests of affected groups in society. Municipalitiesand local administrative departments can play a keycoordinating role. Providing for a high degree of localresponsibility could help to ensure that availableknowledge on coping strategies on the ground is usedefficiently, that affected groups in society are appro-priately involved in planning and decision-makingprocesses, and, in doing so, that coastal managementsystems are accepted by the local population(SEEDS, 2005; WCDR, 2005; Box 3.4-2).

There is still a considerable degree of catching upto do in order to integrate information on the poten-tial impact of climate change systematically intoimplementation of coastal management systems.Despite sound scientific findings regarding thepotential consequences of climate change, there hasbeen scant political effort so far to devise adequatestrategies for action.

Against this background, the German FederalGovernment’s national strategy for integrated man-agement of German coastal areas is laudable (Bun-desregierung, 2006). This strategy takes into accountthe many different players involved and bringstogether the competing interests of protection andutilization of Germany’s coastal areas under a single,integrated concept. It certainly emphasizes climate

change as a major component in the long-term orien-tation of precautionary planning at regional level.However, in view of the gravity of the anticipatedconsequences of climate change, it is necessary toimprove the scientific basis for developing this strat-egy further. Measures aimed at adapting to the con-sequences of sea-level rise and extreme weatherevents will need to become the primary focus offuture strategy.

3.4.1.4Future challenges

Two issues relating to implementing adaptationstrategies need to be emphasized here: the signifi-cance of proactive measures and the special chal-lenges of implementing adaptation strategies indeveloping countries.

Early warning systemsRisk management plans encompass both proactiveand reactive components of adaptation. Proactivecomponents are particularly important for cost-effective adaptation design, because they prevent orat least reduce the chance of a risk translating into adisaster. This is true particularly with regard to sud-den onset hazards. In the past, priority in financingadaptation strategies was given to reactive measuressuch as financing reconstruction of damaged infra-structure in the wake of a natural disaster (WCDR,2005). What appears to be needed, therefore, is areorientation of funding resources combined with ashift in priorities when choosing appropriate adapta-tion strategies. At the World Conference on DisasterReduction (WCDR) in 2005 in the Japanese city ofKobe, it was decided that 10 per cent of funds hith-erto used for ex-post measures in the aftermath ofnatural disasters should be diverted into preventivemeasures over the next ten years (WCDR, 2005;Münchener Rück, 2005a). The significance of proac-

The procedure followed in the KRIM project provides aguideline for handling climate change-induced uncertain-ties with regard to coastal management and shows how theeconomic future of coastal regions could be forecast andplanned. There is still a great need for further research inthis area, however: (1) to explore scenarios involving higherrises in sea level; (2) to extend existing insights regardingregional losses and the costs of different preventativestrategies (e.g. for the managed retreat option); (3) toinclude other sections of coastline in this analysis. It willultimately require a great many small-scale studies of thistype in order to be able to make more reliable supra-regional predictions regarding the financial impacts of cli-mate change, and to be better able to work out the possibleoptions for action.

The cases described show that in these instances, wherethe rise in sea level is clearly below the WBGU guard rail onsea-level rise, the problems can probably be overcome bymeans of appropriate adaptation measures. Unfortunately,a rise in sea level of more than 1m was not studied for theseregions; in many places, successful adaptation in the eventof such a high sea-level rise would no longer be possible atan acceptable cost. In developing countries, meanwhile, theproblem of financial feasibility would arise even with thescenarios and strategies presented above; the adaptationmeasures discussed above, therefore, do not representstrategies that are universally applicable.

Sources: Daschkeit and Schottes, 2002; Mai et al., 2004;Elsner et al., 2005


tive measures is underlined by the plan adopted bythe WCDR to promote an International Early Warn-ing Programme (IEWP).The IEWP is aimed at iden-tifying and closing existing gaps as regards earlywarning (UN ISDR, 2005c). Key elements forimproving early warning systems include developingnational, integrated risk reduction strategies, capac-ity-building in the field of risk management andimproving technical equipment and training. In addi-tion, strategies are to be developed to improve com-munication of warnings to affected communities.Early warning therefore comprises a range ofaspects, from technical capacity to preparatory mea-sures at municipality level. To date, however, thislinking of planning and precautionary measures withadequate response strategies has often been flawed.In future this deficit in existing systems is also to beeliminated. In order to achieve the goals set out inKobe, international cooperation is required particu-larly in the area of data exchange, dissemination ofwarnings and developing institutional structures. Atthe present time there is a particularly urgent need toraise governments’ awareness of the problem andestablish priorities for developing appropriate strate-gies of risk reduction.

Special challenges in developing countries Climate change will have a major impact on develop-ing countries in particular. These countries accountfor 97 per cent of fatalities from natural disasters(Freeman et al., 2003). Damage resulting from nat-ural disasters is a considerable impediment to eco-nomic development in these countries. Adaptationtherefore has particular significance for theseregions. Technical know-how, appropriate institu-tions and especially financial resources are lacking,however, to enable the necessary measures to be putin place. There is broad consensus in the interna-tional community that support should be given tohelp developing countries cope with the impact of cli-mate change. In Article 4, para. 3 of the FrameworkConvention on Climate Change, the Parties to theConvention commit themselves to provide financialand technical support to affected countries. In thecontext of the ‘Hyogo Framework for Action’, the 10-year programme of action adopted at the WCDR,this commitment was reiterated (WCDR, 2005). Inaddition, in recent years there has been increasingrecognition of the fact that strategies for adapting tonatural disasters and slow onset hazards need to bemade an integral part of sustainable developmentcooperation (UNFCCC, 1992; UN ISDR, 2005a, b).

3.4.1.5Financing adaptation measures in developingcountries

To provide financial support to enable developingcountries to adapt to the general consequences of cli-mate change, a variety of international funding insti-tutions offer financial transfers at multilateral level.

International fundsIn recent years, international funding bodies havebeen set up to promote adaptation measures indeveloping countries. In the context of the Frame-work Convention on Climate Change, three fundshave been established that provide funding for adap-tation to climate change in general, in other wordsnot specifically related to oceans: the Special ClimateChange Fund (SCCF), the Least Developed Coun-tries Fund (LDCF) and the Adaptation Fund (GEF,2005b).

It is the explicit mandate of the SCCF to providefunding for adaptation projects and technologytransfer. The fund was set up in 2003 to complementthe Global Environment Facility (GEF) with a spe-cific focus on climate change. By late 2004, the vol-ume of funds in the hands of the SCCF in the form ofvoluntary contributions from OECD countries andother industrialized countries totalled US$34.7 mil-lion. SCCF has been in a position to provide effectivesupport for projects since early 2005.

The LDCF gives particular priority to providingsupport for developing countries to formulate andimplement National Adaptation Programmes ofAction (NAPA). NAPAs identify areas where actionrelating to adaptation is most needed. Of theUS$32.5 million already contributed to the fund,US$11 million has already been disbursed for the for-mulation of NAPAs.

The Adaptation Fund, lastly, was set up with theaim of implementing Article 12, para. 8 of the KyotoProtocol. Its primary source of funds is a share in theproceeds of Clean Development Mechanism (CDM)project activities amounting to 2 per cent of the cer-tified emission reductions issued for a project activ-ity. Disbursal of payments from this fund is unlikelyto begin before 2008, that is, before the start of thefirst commitment period under the Kyoto Protocol.While the revenue effect of this de facto taxation ofprevention projects is welcome, its allocation effectmust be viewed with considerable criticism.

In addition to the above, GEF also provides fundsfor projects under its Climate Change Focal Area. Inthis case, however, the focus is on prevention projectsrather than adaptation projects.

58


Efficient use of development cooperationfunds As well as the above funds, international donors pro-vide financial assistance to developing countriesaffected by natural disasters in the context of devel-opment cooperation. In recent years, for example, theshare of funds made available by the World Bank fordealing with the consequences of natural disasterssuch as tropical cyclones has increased markedly,from 3 per cent to 8 per cent of the World Bankportfolio (Freeman et al., 2003). Financial resourcesare thus increasingly being earmarked for projectsnot aimed at fulfilling the original goal of promotingeconomic and social development.

If the aim of international development coopera-tion is to support the development of adaptive cap-acity in developing countries, then assistance mustfocus to a greater extent on preventative strategiesthan has hitherto been the case, e.g., on developingearly warning systems.A partial shift of this sort fromaftercare to hazard prevention takes on added sig-nificance against the background of expected intensi-fication of climate-induced extreme events. In orderto prevent a loss of efficiency, development coopera-tion should be brought into line with the policies ofthe special adaptation funds described above.

At the same time, while ironing out the issue offinancing adaptation measures, it is important not tolose sight of the actual goal of development coopera-tion. Economic and social development in itselfremains the best form of adaptation strategy, becauseit generally increases the adaptive capacity of adeveloping country and thereby reduces its vulnera-bility to the impacts of climate change (Schelling,1992).

Complementary instruments: Prioritizingmicro-insurance The funding required for adaptation measures can-not be quantified in any robust manner due to a lackof even moderately reliable damage estimates (Sec-tion 3.2). It can nevertheless be assumed that theabove-mentioned financial resources will not be suf-ficient and that it would consequently be sensible tosecure funding for adaptation measures in the broad-est possible manner. For this reason, new fundingmechanisms should be considered alongside existingfunding instruments and reallocation of currentlyavailable resources (WBGU, 2002).

Another means is to promote micro-insurance inorder to disperse the individual risk of hardship; incountries with low per capita incomes, this takes onadded significance. Micro-insurance aims to provideinsurance protection at extremely low premiums forhouseholds and small businesses with a low, and in

some cases irregular, income and to increase avail-able financial resources in the event of losses occur-ring. Micro-insurance, therefore, is not concernedwith the national or international level, but is aimedat protecting individual assets (Münchener Rück,2005b).

Micro-insurance experience already exists insome areas where individual risks occur indepen-dently of each other, e.g. risks relating to illness oraccidents (Brown and Churchill, 1999, 2000; Ahmedet al., 2005; Cohen et al., 2005). Case studies carriedout in India, Kenya or Uganda show that life insur-ance and health insurance in particular are alreadybeing used successfully (Brown and Churchill, 1999,2000; Athreye and Roth, 2005). Micro-insurance forrisks relating to natural disasters, on the other hand,is still being piloted. Applying micro-insurance tonatural disasters is particularly difficult because largenumbers of people are usually affected and the indi-vidual risk of loss to local policyholders thus dependson the risk for all the others. As a result, demands onthe insurer are very high in the event of loss occur-ring, and may even exceed the insurer’s capital stock.If the insurance provider opts for increasing his cap-ital stock or reinsuring as a means of solving theproblem, his capital costs increase and this isreflected in the price of the insurance policy. In thesecircumstances, many households and businesses withlow incomes will ultimately forego private insurancealtogether.

In order to be able put a reasonably-priced andeffective insurance product within their reachdespite the difficulties, existing micro-insurance sys-tems for independent risks could be extended tocover losses arising as a result of natural disasters.The costs of insurance cover are kept low by devel-oping effective institutional capacities and ‘bundling’policyholders in groups and municipalities. In addi-tion, governments could make it compulsory to takeout insurance against natural disasters. This wouldenable a large number of policyholders to berecruited swiftly and achieve a broad geographicaldistribution of policyholders, which would greatlyreduce the problem of correlated risks of individuallosses. The question of whether compulsory insur-ance of this sort would be a sensible option – espe-cially in countries where social insurance systems arestill inadequate – should be investigated in the con-text of future research.

In order to ensure that providers of insurance fornatural disasters operating at national or regionallevel are successful in the long term, it is importantthat they are linked to the international capital mar-ket. For example, against payment of a premium,reinsurance companies act as ‘insurers of the insur-


ers’, assuming a proportion of the insuranceprovider’s risk. Risks are thus spread more broadlyand insurers are freed from the risk of facingextremely high payouts.

Micro-insurance programmes should be activelypromoted by governments (public co-financing):alongside establishing the necessary legal frame-work, providing financial support might also be con-sidered in the early stages, especially to develop thenecessary institutional infrastructure, for example inthe context of public-private partnerships and incooperation with development organizations (Linne-rooth-Bayer and Mechler, 2005).

3.4.2The adoption of provisions governing loss ofterritory in international law

Adaptation strategies raise a number of legal issuesas well. With steadily rising sea levels, it is likely thatmanaged retreat will be the only option in manycases. In particular, national territories may well belost completely or partially as a result of flooding,with people being forced to abandon settled areas. Interms of international law, this poses various chal-lenges which relate, firstly, to the resettlement of thepeople displaced by sea-level rise, and secondly, thequestion of financial compensation in cases whenstates which are affected by the impacts of climatechange-induced sea-level rise have not contributedsignificantly to its causes.

3.4.2.1Reduction in the size of national territory

If a state’s territory shrinks as a result of sea-levelrise, this does not have any specific implications interms of international law, aside from the issue ofcompensation (see Section 3.4.2.4). According to therelevant provisions of international law, in such a sce-nario, the constituent national territory is simplyreduced in size. In individual cases, however, it maybe necessary to amend specific commitments arisingunder international law, primarily those relating tothe territory which is now submerged. In general, therelevant provisions of international law supply satis-factory solutions to the legal problems which can beanticipated here. It must be borne in mind that areduction in the size of a state’s national territorymay result in a shift in the boundaries of its maritimejurisdiction as well, if the points used to positionthem have changed.

3.4.2.2Submersion of (island) states

According to current knowledge, the survival ofisland states lying only a few metres above sea levelis in acute jeopardy as a result of climate change-induced sea-level rise (CSD, 2004). These islandstates include the Maldives, an island group lying nomore than 2m above sea level, and the Tuvalu, Kiri-bati and Tonga island groups, which are located oncoral reefs. These small island states, which are alsodeveloping countries (Small Island DevelopingStates – SIDS), have formed a community of interestwhich is making its presence felt as a political alliancein the international negotiations on the UnitedNations Framework Convention on Climate Change(UNFCCC) (Burns, 1997; Neroni Slade, 2001).Admittedly, the SIDS (along with countries with low-lying coastal areas) have been given special consider-ation in the UNFCCC; for example, Article 4, para. 8(a) and (b) calls for consideration to be given toactions, including funding, insurance and the transferof technology, which may be necessary to meet thespecific needs and concerns of these countries. How-ever, this vague reference comprises the full extent ofthe special consideration of island states contained inthe UNFCCC. Indeed, Article 4, para. 8 of theUNFCCC defines the specific needs of other cate-gories of developing countries in such broad termsthat almost any developing country Party could claimto be particularly vulnerable in some way. In otherwords, no specific rights for the island states can bederived from these provisions of the Convention.Theisland states are not mentioned specifically in theKyoto Protocol. In the supplementary agreementsadopted by the Parties to operationalize the KyotoProtocol, notably the Marrakech Accords, the needsof the island states are emphasized repeatedly, butthis has yet to result in the adoption of any institu-tional or other specific provisions.

Other regional or global agreements, especially inthe law of the sea, also fail to recognize, in any legallymeaningful way, the status of island states as coun-tries with special ecological or other problems. Thesame applies to the United Nations Convention onthe Law of the Sea, even though islands play a keyrole in this Convention as a maritime geographicalcategory of importance in determining maritimezones and related sovereign rights (Jesus, 2003).

From the perspective of international law, theexistence of a national territory is a constituent ele-ment of the state, which means that submersion of astate’s territory could result in its extinction. Nordoes international law currently grant any entitle-ment to the allocation of any kind of ‘replacementterritory’, although this would be possible in political

60


terms. However, experience in the Middle East, notleast, has shown that the creation of a state or newnational territory may trigger considerable potentialfor conflict, especially given that hardly any unsettledterritories are now available for consideration.

3.4.2.3Dealing with ‘refugees from sea-level rise’

If a state is submerged, its citizens become stateless.‘Refugees from sea-level rise’ will probably seekrefuge in neighbouring countries, perhaps greatlyexceeding these countries’ absorption capacities.WBGU therefore considers that formal provisionsare required to regulate the legal status of these peo-ple.

WBGU recommends that the adoption of rele-vant provisions under international law be guided bythe following principles. Basic provisions shouldestablish the affected population’s right to regulatedrefuge/resettlement. This raises the question of theobligations which would thus arise for potential hostcountries, whereby a distinction must be madebetween the practical reception of the refugees andthe covering of costs. From a humanitarian perspec-tive, the best option is for refugees to be received bycountries in the geographical vicinity of, or with spe-cific links to, the submerged state. The refugeesshould have a say in choosing their new living envi-ronment; forced resettlement should be avoided asfar as possible. At the same time, however, an alloca-tion formula should be developed in a process involv-ing the wider international community, in order toensure that individual host countries’ capacities arenot overstretched. Fair and efficient burden-sharingrequires that the costs of receiving the refugees beallocated according to the ‘polluter pays’ principle.The allocation formula should thus be guided by theprinciple of common but differentiated responsibilityenshrined in international law. This means that theheaviest burden must be borne by those countrieswhich are making the largest contribution to globalgreenhouse gas emissions and which also have thegreatest financial resources at their disposal (Princi-ple 7 of the Rio Declaration, Article 3, para. 1 andArticle 4, para. 1 of the UNFCCC; Kellersmann,2000; Stone, 2004). It is important to bear in mind thatthe issue of sea-level refugees is universal, for it arisesnot only when an individual state is submerged butalso when major climate change-induced floodingand devastation occur in states which continue toexist.

The development and application of the relevantlegal provisions may prove to be problematical inpractice, however. How can refugees who have lost

their living environment as a result of climate change,making them dependent on assistance from others,be distinguished from other refugees? And how canthe fundamental problem of causality be resolved?After all, hurricanes or extreme weather conditionswhich trigger refugee flows may not necessarily becaused by climate change but may simply be theresult of the natural variability of the climate system(Section 3.1.2; Stone and Allen, 2005). Solutions tothese problems must be found when formulatinglegal provisions governing the treatment of sea-levelrefugees. Against this background, WBGU recom-mends a significant increase in research in this area,especially the analysis and exploration of fair andeffective burden-sharing systems.

A further difficulty arising in this context is that‘environmental refugees’ do not fit into any acceptedcategory in international refugee and migration law(GCIM, 2005).According to the Convention relatingto the Status of Refugees (Geneva Refugee Conven-tion), the term ‘refugee’ only applies to persons per-secuted for reasons of race, religion, nationality, [or]membership of a particular social group or politicalopinion. It does not create any specific obligationsunder international law for the treatment of ‘sea-level refugees’. In WBGU’s view, this gap in interna-tional refugee law must be closed. One option is toestablish bilateral agreements, e.g. with neighbourstates, or to adopt a multilateral agreement. Thisraises the question whether the existing conventions,especially the Refugee Convention, can be amendedappropriately without renegotiating the definition of‘refugee’ itself, or whether the conclusion of a sepa-rate convention would be more appropriate. In linewith the non-refoulement principle, persecuted per-sons may not be deported to a country where theymay be subjected to torture or inhumane treatment.By the same token, states should undertake not toreturn sea-level refugees to their country of origin ifclimate change has rendered the conditions of life inthese countries unsustainable, i.e. if the living condi-tions are incompatible with human dignity or basiceconomic survival cannot be guaranteed. The scopeof such a new norm must therefore extend beyondthe specific problem of sea-level refugees to encom-pass other forms of environmentally related migra-tion as well.

3.4.2.4Compensation for loss of land

Compensation issues play a key role in relation to theloss of territory and the submersion of island states.A distinction must be made between various scenar-ios here.


In cases when only the national level is affected,i.e. the damage is sustained by private individualsthrough the loss of their property or its value, or lossof income, national law applies; such cases are notrelevant to this report. However, possible interna-tional conventions may have an impact on privateactors if, for example, a state passes the responsibilityfor collecting the resources to cover internationalagreed compensation payments through taxes andlevies to the private sector.

What is relevant, however, is whether and to whatextent the international community or other individ-ual states have an obligation to pay compensation ifa country sustains damage directly or indirectly as aresult of sea-level rise. According to current interna-tional law and practice and prevailing opinion, nosuch obligation exists: even though the problem ofrising sea levels is rarely caused by the affected islandor coastal states but is primarily due to greenhousegas emissions in the industrialized and newly indus-trializing countries, an obligation to pay reparationsor damages does not arise under current interna-tional law. The background to this issue is the prob-lem of the cumulative effects of certain types of con-duct – a question which has yet to be satisfactorilyresolved in international law – and the causal links,which are sometimes difficult to establish. As inter-national law stands, the ban on causing major trans-boundary environmental injury, recognized in cus-tomary international law, thus does not apply(Epiney, 1995; Beyerlin, 2000; Wolfrum, 2000; Sands,2003). Nonetheless, cause and effect have been estab-lished in many instances, and there is no doubt thatclimate changed-induced sea-level rise is presentingsome developing countries with problems which theylack the financial resources to cope with unaided.

Against this background, WBGU recommendsthe conclusion of an international convention whichwould oblige the industrialized countries in particu-lar to guarantee adequate funding for an internation-ally administered compensation fund. Fundingwould be disbursed from this fund to countries par-ticularly affected by rising sea levels.A country’s con-tribution commitments should be weighted accord-ing to the greenhouse gas emissions it produces, sothat payments can be regarded as compensation for acountry’s actual contribution to climate-related dam-age (Section 3.4.1.5). Once this compensation fundhas been established as a means of providing assis-tance to the affected states, it could also take on arole in burden-sharing within the international com-munity, e.g. managing the reception of refugees flee-ing from sea-level rise and the payments made tohost countries (Section 3.4.2.3).

Utilizing the mechanisms for the transfer of finan-cial resources and technology established for the cli-

mate regime might also appear, at first sight, to be aviable option. For example, the Mauritius Strategyfor the Further Implementation of the Programme ofAction for the Sustainable Development of SmallIsland Developing States (para. 78(a)) posits, in thecontext of climate change adaptation and sea-levelrise, that strategies can be developed with supportfrom the Least Developed Countries Fund and theSpecial Climate Change Fund set up within theframework of the United Nations Framework Con-vention on Climate Change. The key objection hereis that such support cannot be regarded as genuinecompensation for climate-induced damage.A furtherpossibility is for the United Nations CompensationCommission to take action in this area; for example,it recently adjudicated compensation for environ-mental damage caused in the 1990–1991 Gulf War(Sands, 2003). However, this particular instrument isnot precise enough to pay targeted compensation forthe damage caused by climate changed-induced sea-level rise.An existing body could at best be entrustedwith the task of administering the separate compen-sation scheme outlined above.


Hurricane formation and strengthThe links between hurricane activity and globalwarming need to be researched more thoroughly,both through further analysis of data gathered frompast developments and by modelling the futuredevelopment of the hurricane climate, includingpotential threats to areas not affected previously(South America, southern Europe).

Extent and rate of sea-level riseThe greatest uncertainty surrounding future sea-level rise concerns the behaviour of continental icesheets in Greenland and Antarctica. To reduce thisuncertainty, there is a need to gain an improvedunderstanding of ice dynamics; major progress isneeded in continental ice modelling. These activitiesinclude researching the stability of the ice shelves aswell as their interplay with continental ice. Furtheruncertainties surround ocean dynamics, especiallythe intensity of ocean mixing. Such dynamics greatlyinfluence sea levels. There is a need to improve theircharacterization within global climate models.

Global potential for damage caused by sea-level riseThe issue of ‘dangerous sea-level rise’ forms a sub-setof the wider question of ‘dangerous climate change’and must be answered quantitatively if possible. To

62


do so, there is a need to aggregate globally the health,socio-economic and ecological consequences associ-ated with various scenarios (x metres rise in y years).Present assessments are not robust in this respect.They must be replaced by a new generation of impactanalyses. This could produce a more precise defini-tion of the provisional absolute guard rail proposedby WBGU (maximum of 1m sea-level rise).

Vulnerability of coastal megacities indeveloping countriesClimate change and urbanization are dominanttrends of global change. The interplay of the twotrends in the major coastal cities of the developingworld could cause an almost unmanageable situa-tion, particularly if the arsenal of responses is limitedby social, economic and institutional deficits.There isan urgent need to conduct interdisciplinary studies inorder to assess the severity of the problems for par-ticularly critical megacities such as Lagos, Mumbai orHavana.

Regional portfolio strategies for coastalmanagementThe dramatic geophysical impacts of climate changeupon coastal zones (which will arise even if vigorousmeasures are taken to reduce global greenhouse gasemissions) mean that the traditional approaches tocoastal management need to be revisited. There is aparticular need to determine the priority given to thevarious strategic elements of protection, managedretreat and accommodation. To be able to conductsuch an assessment, types of cost-benefit analysismust be developed that take account of the novelpotential for damage. At present such analyses haveonly been carried out for limited sections of coasts,e.g. in Great Britain. There is an urgent need to con-duct an integrated re-appraisal of robust and effec-tive portfolio strategies for German coasts.

‘Sea-level refugeese’: Legal andinstitutional aspectsThe threats presented to coastal regions and thepotential destruction of entire state territories by cli-mate-induced sea-level rise generate a novel migra-tion problem whose legal dimensions have yet to beexplored. There is a particular need for research onhow to shape provisions under international law withrespect to the reception of ‘sea-level refugees’, thepayment of compensation, and burden-sharing inline with the ‘polluter pays’ principle. To resolve thelegal problems, it is also very important to makeprogress in the scientific attribution of damage or ter-ritorial loss arising as a consequence of human-induced climate change. There is also a need to con-duct operative appraisals, for instance to evaluate the

capacity of the existing United Nations institutions tocope with refugee flows, especially given that needswill presumably grow exponentially in the future.

Ocean acidification

4.1Chemical changes in seawater

4.1.1CO2 input

The oceans hold around 38,000 gigatonnes of carbon(Gt C). They presently store about 50 times moreCO2 than the atmosphere and 20 times more than theterrestrial biosphere and soils (Fig. 4.1-1). However,the ocean is not only an important CO2 reservoir, butalso the most important long-term CO2 sink. Drivenby the difference in the partial pressure of CO2

between the atmosphere and seawater, a portion ofthe anthropogenic CO2 dissolves in the surface layerof the sea and, over periods ranging from decades tocenturies, is finally transported into the deep sea byocean currents.

There has already been a demonstrable increasein CO2 concentrations in the upper layer of the seaover recent decades (Sabine et al., 2004) that can beattributed to the proportional rise of CO2 in theatmosphere. The ocean is presently taking up 2Gt ofcarbon annually, which is equivalent to about 30 percent of the anthropogenic CO2 emissions (IPCC,2001a). Altogether, between 1800 and 1995, theoceans have absorbed around 118Gt C ± 19Gt C.That figure corresponds to about 48 per cent of thecumulative CO2 emissions from fossil fuels (includ-ing cement production), or 27–34 per cent of the totalanthropogenic CO2 emissions (including those fromland-use changes; Sabine et al., 2004). The anthro-pogenic CO2 signal in the sea can be traced, on theaverage, to a water depth of approximately 1000m.Due to the slow mixing of ocean layers it has not yetreached the deep sea in most parts of the ocean. Inthe North Atlantic, however, due to the formation ofdeep water there, the anthropogenic CO2 signalalready extends down to 3000m.

Fossilfuels

> 6,000

92 90

6

12260

60

1Net destructionof vegetation

Rivers 0.4 DOC 0.4 DIC

Burial 0.1

+ 3 Gt C/a

Landplants560

Soils1,500

Atmosphere800

(100–100,000 Jahre)37,000

Intermediate and deep ocean

Surface layer 900

(3–7 years)

(5–10,000 years)

(50 years)

Marine sediments 30 million

Total ocean+ 2 Gt C/a

FFiigguurree 44..11--11Overview of the globalcarbon cycle. Values for thecarbon reservoirs are givenin Gt C (numbers in bold-print). Values for theaverage carbon fluxes aregiven in Gt C per year(numbers in normal-print).Mean residence times are inparentheses. Flux into soilsamounts to around 1.5Gt Cper year. DOC = dissolvedorganic carbon, DIC =dissolved inorganic carbon.Sources: adapted afterSchlesinger, 1997 andWBGU, 2003.Numbers expanded andupdated for ocean and fossilfuels: Sabine et al., 2003;marine sediments: Raven etal., 2005; atmosphere:NOAA-ESRL, 2006

4

4 Ocean acidification

In the atmosphere CO2 behaves chemically neu-tral, that is, it does not react with other gases, but itcontributes to climate change through its stronginteraction with infrared radiation. But in the oceanCO2 is chemically active. Dissolved CO2 contributesto the reduction of the pH value, or an acidificationof seawater. This effect can already be measured:since the onset of industrialization the pH value ofthe ocean surface water has dropped by an averageof about 0.11 units. This is equivalent to an increasein the concentration of hydrogen ions (H+ ions) byaround 30 per cent. Starting from a slightly alkalinepre-industrial pH value of 8.18 (Raven et al., 2005),the acidity of the ocean has thus increased at the sur-face. The various IPCC emission scenarios indicatethat if the atmospheric CO2 concentration reaches650ppm by the year 2100, a decrease in the averagepH value by 0.30 units can be expected compared topre-industrial values. With an atmospheric concen-tration of 970ppm, the pH value would drop by 0.46units. But if the CO2 in the atmosphere can be limitedto 450ppm, then the pH reduction will only amountto 0.17 units (Caldeira and Wickett, 2005).

4.1.2Change in the carbonate budget

The carbon stored in the seas occurs in differentchemical forms. A small part is stored in the bios-phere and in organic compounds, but the greatestpart by far is contained in inorganic compounds,which are referred to as DIC (dissolved inorganiccarbon). Of these compounds, however, only 1 percent is directly dissolved CO2, 91 per cent occurs asbicarbonate (HCO3

–), and 8 per cent as carbonate(CO3

2-). The relationship of these three compoundscan be represented by the equilibrium equation:

CO2 + H2O + CO32-

↔ 2 HCO3–

The relative proportions of these carbon compoundsreflect the pH value of the water (Fig. 4.1-2). OnlyCO2 can be exchanged with the atmosphere.Throughthe uptake of CO2 the partial pressure of CO2

increases in the seawater, and at the same time theequilibrium shifts in favour of bicarbonate and to thedetriment of carbonate.

Due to the uptake of anthropogenic CO2, the car-bonate concentration in the ocean surface layer hasalready dropped by 10 per cent compared to the pre-industrial level (Orr et al., 2005).

The saturation of seawater with carbonate ions isespecially important for marine organisms that buildtheir shells or skeletons with lime (calcium carbon-

ate, CaCO3: Section 4.3.2). Calcium carbonate occursin marine organisms primarily in the forms of arago-nite and calcite, which differ in their crystal structures(Table 4.3-1). Seawater is supersaturated withrespect to the more easily dissolved aragonite whenthe carbonate concentration lies above 66µmol perkilogram. If it falls below this value the aragoniteformed by the organisms dissolves in the water – thisis referred to as aragonite undersaturation. Becauseof the increasing solubility of calcium carbonate withdecreasing temperature and increasing pressure, thedeeper layers of the sea are, as a rule, undersaturated,that is, sinking CaCO3 dissolves in the water atgreater depths. The boundary between the undersat-urated and super-saturated layers is referred to as thesaturation horizon.

The present carbonate concentration in the seasurface layer varies among regions: the highest con-centrations (averaging 240µmol per kilogram) occurin the tropics, while values in the Southern Oceanaverage only 105µmol per kilogram (Orr et al., 2005).With progressive CO2 input into the sea, therefore,the marine organisms in the Southern Ocean are thefirst to be threatened by aragonite undersaturation(Section 4.3.2). Orr et al. (2005) calculate the possiblefuture development of the carbonate concentrationof the Southern Ocean for various emission scenar-ios. According to these calculations, under a ‘busi-ness-as-usual’ scenario it could already be undersat-urated with respect to aragonite by the middle of thiscentury (Fig. 4.1-3). With an atmospheric CO2 con-centration of approximately 600ppm or more, thegreater part of the surface layer of the SouthernOcean would be undersaturated. But even beforethis threshold is reached the saturation horizon driftsupward, that is, the upper layer of the sea that is

66

HCO3-

CO32-

CO2

pH-rangetoday

Expected

change

1

0.1

0.01

0.00165 74 8 9 10 11

pHacidic basic

Pro

por

tion

of t

he c

once

ntra

tions

FFiigguurree 44..11--22Carbonate system of seawater. Relative proportions of thethree inorganic components CO2, HCO3

–und CO3

2-. The blueshaded area shows schematically the pH range that occurs intoday’s ocean. The arrow shows the expected shift of theaverage pH value when the atmospheric CO2 concentrationreaches about 750ppm.Source: Raven et al., 2005

67Future development of the oceans as a carbon sink 4.2

supersaturated with respect to aragonite becomesthinner, and CaCO3 formation becomes more diffi-cult. Individual parts of the surface layer would beaffected even earlier. With respect to the less-solublecalcite the surface layer remains saturated despitehigher CO2 concentrations, but the calcite saturationhorizon also shifts upward. With the displacement ofthe saturation horizon the conditions for biogenicCaCO3 formation change, which can have consider-able consequences for marine organisms and ecosys-tems (Section 4.3).

4.1.3Special role of CO2

The acidification of the sea is an effect that can beexclusively attributed to the CO2 increase in theatmosphere. In this it is different from climatechange, which is caused by the radiative effect ofatmospheric CO2 increase, but also of the increase ofmethane, nitrous oxide and several other radiativelyactive gases. With respect to climate change, calcula-tions are often made in terms of CO2 equivalents,that is, the radiative forcing attributable to the vari-

ous gases is recalculated to the corresponding forcingof CO2.The argument is that for climate protection itdoes not make any difference whether the radiativeeffect is caused by CO2 or by any other emittedgreenhouse gas. But this is not true for the effect ofocean acidification. To protect the oceans, reducingCO2 emissions is relevant for two reasons: to limitboth global warming and ocean acidification.

Acidification is, above all, a consequence of therapid increase of the quantities of CO2 in the ocean.With a slow input of CO2, as has repeatedly occurredin the Earth’s history (such as the end of the last iceage when the CO2 concentration rose by 80ppm overa period of 6,000 years), or in climate epochs withelevated CO2 concentrations (around 100–200 mil-lion years ago) the CO2 mixes down into the deepsea, where a slow dissolution of carbonate sedimentscounteracts the acidification. In such constellationsthe pH value of the sea remains almost constant(Raven et al., 2005).

4.2Future development of the oceans as a carbonsink

As discussed in Section 4.1, the oceans are the mostimportant net sink for CO2. Without oceanic uptakeof anthropogenic CO2, the relative CO2 concentra-tion in the atmosphere would lie more than 55ppmabove the present level (Sabine et al., 2004). Thefuture development of the oceans as a CO2 sink willtherefore determine in large part how stronglyanthropogenic CO2 emissions are reflected as anincrease in the atmospheric concentration of carbondioxide. Over the long term, that is, a period of sev-eral centuries (in which mixing takes place through-out the world’s oceans), the ocean can take up about65–80 per cent of the anthropogenic CO2, dependingon the total quantity of carbon emitted. At evenlonger time scales this proportion increases to 85–92per cent due to the dissolution of carbonate sedi-ments (Caldeira, 2005). In the coming decades andcenturies, however, only a portion of this great sinkpotential can be effective: the limiting factor is thetransport of carbon taken up at the surface into thedeeper ocean layers. In fact, the oceans have so faronly absorbed 30 per cent of the amount of anthro-pogenic carbon that they could take up over a longtime period at present atmospheric concentrations(Sabine et al., 2004).

The great importance of the ocean as a sink is notapplicable to the other greenhouse gases regulatedby the Kyoto Protocol: the strongest sink for methaneas well as for HFCs, for example, is the chemical reac-tion with the hydroxyl radical OH in the lower

A2

A1 FI B1A1 B B2A1 T IS 92a

Aragonite saturation

Calcite saturation

IPCC Scenarios1,000

800

600

400

20020202000 2040 2060 2080 2100

Year

Atm

osp

heric

CO

2 [p

pm

]

120

80

100

60

40

20

020202000 2040 2060 2080 2100

Year

Car

bon

ate

in s

urfa

ce o

cean

[µm

ol/k

g]a

b

FFiigguurree 44..11--33Projections of different CO2 concentrations (a) and theireffects on the carbonate budget of the Southern Ocean (b).The variation according to various IPCC scenarios is shown.Source: Orr et al., 2005


atmosphere, while N2O is destroyed primarily in thestratosphere by UV radiation from the sun. Theindustrial gases PFCs and SF6 do not decay until theyare above the stratosphere. It is worth noting, how-ever, that the sea is an important source of N2O,whose future development in response to climatechange is unclear.

Before industrialization the ocean was at a state ofnear equilibrium, and not a CO2 sink.At its surface itgave off around 0.6Gt C annually to the atmosphere,while at the same time approximately the sameamount of carbon entered the ocean from the terres-trial biosphere (and therefore ultimately from theatmosphere) in the form of organic matter flowing infrom rivers (Watson and Orr, 2003). The proportionof atmospheric CO2 did not change under these con-ditions, remaining constant over millennia at around280ppm. The reason for the present function of theocean as a sink is the anthropogenic perturbation ofthe carbon cycle: when the CO2 concentration of theatmosphere increases, the ocean takes up CO2 untilthe partial pressures of the surface water and theatmosphere are in equilibrium. Since the beginningof industrialization the atmospheric CO2 concentra-tion has risen almost exponentially. This has causedan annual increase in the CO2 uptake by the oceanssince that time, in quantities almost proportional tothe atmospheric CO2 concentrations, as model stud-ies indicate (Gloor et al., 2003). For various reasons,however, this cannot be carried over into the future,which will be discussed below.

When one compares the quantities of CO2 takenup by the ocean with anthropogenic emissions, theefficiency of the ocean sink appears to be fallingalready: Sabine et al. (2004), based on an analysis ofobservational data, show that from 1800 to 1994 theocean absorbed 28–34 per cent of the anthropogenicemissions, while from 1980 to 1999 this value wasonly 26 per cent. Due to the large uncertainty in thedetermination of the global carbon balance, thisdecrease is not statistically significant, but on thebasis of known geochemical processes it is also notunexpected.The more CO2 that has been taken up bythe ocean, the lower the carbonate concentration inthe surface layer becomes (Section 4.1.2). Thisdecreases its capacity to take up additional CO2.Modelling studies show that the relative CO2 uptakeby the ocean (that is, the proportion of anthro-pogenic emissions absorbed by the ocean in thecourse of a few decades) is reduced by this effect byseveral per cent when an atmospheric CO2 concen-tration of 450ppm is reached. At 750ppm of CO2 inthe atmosphere the relative CO2 uptake falls by asmuch as 10 per cent (Le Quéré, personal communi-cation).This geochemical effect is fully considered inmodels of the carbon cycle and is therefore rarely

expressly discussed (Gruber et al., 2004). This effectis also active in the extreme long term, that is, timeperiods in which the ocean completely mixes, so thatthe proportion of anthropogenic CO2 emissionsremaining in the atmosphere continues to increase asmore CO2 has been emitted.

Climate change resulting from greenhouse gasemissions further affects the capacity of the oceansink: the solubility of CO2 in seawater decreases withrising temperature. Through this effect, by the end ofthis century the cumulative CO2 uptake could fall by9–14 per cent of what it would be without a tempera-ture change (Greenblatt and Sarmiento, 2004). Thiseffect is well-understood; the uncertainty predomi-nantly results from the uncertainty of the degree ofexpected temperature change.

A further effect of climate change is an increasingocean stratification, that is, the vertical mixing will bereduced. This has a number of complex effects. Forone, the transport of carbon-enriched surface waterto greater depths as well as the transport of carbon-depleted water to the surface will be weakened,resulting in an overall decrease of the sink effect ofthe ocean. For another, there could be changes in bio-logical productivity through altered nutrient avail-ability. Biological productivity is of great importancefor the carbon balance of the ocean surface layer:CO2 is taken up by marine organisms through photo-synthesis and incorporated into organic substance;dead organisms sink and then decay in differentwater depths. Part of the released nutrients and car-bon return to the surface through vertical mixing, butthe net export to the deep sea is considerable. Tengigatonnes of carbon are transferred annually by this‘biological pump’ from the ocean surface layer to thedeep sea.The combined effect of increased stratifica-tion and altered biological productivity on the sinkeffect of the ocean is highly uncertain. Greenblattand Sarmiento (2004) give a range of -2 per cent(decreased sink function) to +10 percent (increasedsink function) for the change in cumulative CO2

uptake through this effect by the end of the century.Many of the effects discussed are still difficult to

quantify, but it is likely that climate change will con-tribute to a considerable overall weakening of theefficiency of the sea as a carbon sink.According to anoverview based on various modelling studies byGreenblatt and Sarmiento (2004), the cumulativeCO2 uptake by the ocean could be 4–15 per centlower by the end of the century due to the climate-related influences discussed above (temperature rise,increased stratification, and biological effects) than itwould be without these. This attenuation of the CO2

uptake has to be added to the geochemical effectsthat already lead to a weakening of the relative sinkwith a similar order of magnitude.

68

69Effects of acidification on marine ecosystems 4.3

As already indicated, biological processes repre-sent the greatest uncertainty in estimating the futuredevelopment of the ocean sink. These biologicalprocesses include the impacts of anthropogenicinterference with the atmosphere and ocean acidifi-cation on marine primary production, the biologicalpump and calcification (Section 4.3.5). A weakeningof the ocean sink due to changes in the wind-drivenrise of water at the equator (‘equatorial upwelling’)is a further aspect under debate (Winguth et al.,2005). In addition, non-linear events that are difficultto predict such as a strong decrease in oceanic con-vection or in the thermohaline circulation, or biolog-ical regime shifts (Section 2.2.1) could have a consid-erable influence.

In summary it can be stated that, with increasingatmospheric CO2 concentrations, the proportion ofanthropogenic CO2 emissions taken up by the oceanwill decrease, even if the absolute rate of uptakeincreases (IPCC, 2001a).

4.3Effects of acidification on marine ecosystems

CO2 input into the sea leads to shifts in the carbonatesystem of the seawater and to a decrease in pH value,and thus to acidification of the ocean (Section 4.1.1;Turley et al., 2006). Without counteractive measuresthis change in the carbonate system could reach astate during this century that has probably not beenseen for several million years (Feely et al., 2004).Humans are significantly interfering with the chemi-cal balance of the ocean, and this will not remainwithout consequences for marine organisms andecosystems.

4.3.1Physiological effects on marine organisms

A strong increase of CO2 concentration (hypercap-nia) has many adverse physiological effects that havebeen investigated experimentally on various marineorganisms. Numerous changes in marine organismshave been identified, for example, in the productivityof algae, metabolic rates of zooplankton and fish,oxygen supply of squid, reproduction in clams, nitrifi-cation by microorganisms, and the uptake of metals(for a survey, see Pörtner, 2005). Many of theseexperiments, however, were carried out with CO2

concentrations much higher than what could beexpected in emission scenarios under discussiontoday for the time frame up to 2100. Further studiesare therefore necessary in order to be able to esti-mate the short- and medium-term effects of acidifi-

cation (Section 4.6). From today’s viewpoint it seemsimprobable that marine organisms will suffer fromacute poisoning at expected future CO2 levels (Pört-ner, 2005).

Doubling the present CO2 concentration leads toan increase in the rate of photosynthesis in manyphytoplankton species by about 10 per cent (Ravenet al., 2005). However, the various groups of phyto-plankton exhibit different sensitivities to increasedCO2 concentrations with respect to photosynthesis,which is due to differences in carbon uptake (CO2

versus HCO3–) and a different saturation behaviour

of the photosynthetic rates.The interactions betweenphotosynthesis, primary production of phytoplank-ton, microbial respiration, and the resulting effectson the food web are, however, compounded by anumber of other factors (temperature, light andnutrient supply, disparate feeding risk from zoo-plankton, adaptive processes, etc.). With the presentstate of knowledge no clear conclusions can be drawnregarding the effects of acidification on growth ratesand assemblage compositions of the phytoplankton.

4.3.2Effects on calcifying organisms

Next to photosynthesis, calcification is the mostimportant physiological process influenced by theincrease of CO2 concentration. It has far-reachingconsequences for the ecological function of marineecosystems, and can also have feedbacks on theatmospheric concentration of CO2 and thus on theclimate system (Section 4.3.5).

For their skeletons or shell structure, many marineorganisms use calcium carbonate, which has to beextracted from seawater. This is only possible whilethe seawater is supersaturated with calcium carbon-ate, which is why the increasing CO2 concentrationand falling pH value hampers calcification (Raven etal., 2005). This causes a weakening of the skeletalstructure or – when a level below the saturation con-centration is reached – even their dissolution. Cal-cium carbonate is employed as a construction mater-ial for organisms in different crystalline forms: ara-gonite and calcite are the two most important (Table4.3-1). Organisms that use aragonite for their shellsor skeletons are the first to be adversely affected byacidification, because aragonite dissolves more easilyunder the changing conditions due to its differentcrystal structure.

Acidification has an impact on all marine calcify-ing species, such as certain plankton groups, clams,snails and corals. Echinoderms (for example, starfishand sea cucumbers) are especially threatened,because their calcite structures contain larger


amounts of magnesium and therefore dissolve evenmore easily than aragonite under increased CO2 con-ditions (Shirayama and Thornton, 2005). Althoughcorals are the most conspicuous and well-knownmarine calcifying organisms and are especiallythreatened by acidification as aragonite producers(Section 2.4), they only contribute 10 per cent of theannual global marine carbonate production of0.64–2Gt C (Zondervan et al., 2001).The simulationsof Guinotte et al. (2003) indicate that at an atmos-pheric CO2 concentration of just under 520ppm,which could already be reached by the middle of thiscentury, almost all of today’s warm-water coral reeflocations will barely still be suitable for coral growthbecause of insufficient aragonite saturation (Fig. 4.3-1).

Around three-fourths of the global marine cal-cium carbonate production is carried out by plank-tonic organisms, primarily coccolithophores,foraminifera, and pteropods. Of these, the coccol-ithophores are of particular importance becausethese one-celled primary producers, which can createlarge-area plankton blooms with only a few species,can greatly contribute to the export of calcium car-bonate to the deep sea and thereby play a significantrole in the global carbon cycle (Riebesell et al., 2000;Zondervan et al., 2001; Section 4.3.5). In experimentswith both monocultures and natural plankton com-munities it has been shown that the calcification bycoccolithophores clearly decreases with increasedatmospheric CO2 concentrations (Riebesell et al.,2000; Riebesell, 2004). Pteropods are important com-ponents of the marine food webs in polar and sub-polar latitudes where they form dense populations(up to 1000 individuals per m3) and serve as nutritionfor the upper trophic layers of the food web. In theseregions they are responsible for a significant portionof the export of particulate carbon to greater depths.

When the carbonate saturation in seawater dropsbelow a critical value, it is likely that these animalsare no longer able to form a shell. For importantparts of their habitat in the Southern Ocean anundersaturation with respect to aragonite (underassumptions of the IS92 scenario of the IPCC) is pre-dicted beginning in 2050, so that their distributionarea will be severely limited (Orr et al., 2005; Ravenet al., 2005).

There is great uncertainty about the capacity ofthe organisms to adapt to these changes, as too fewlong-term experiments have been carried out (Ravenet al., 2005; Pörtner, 2005).

4.3.3Ecosystem structure and higher trophic levels

Over the course of this century the expected pHdecreases could have a considerable impact on thecalcifying organisms and thus on the total marinebiosphere (Orr et al., 2005). Simultaneously, consid-erable climate-related warming is expected. The twoeffects are not independent of one another: the CO2

increase, for example, could decrease the tempera-ture tolerance of animals (Pörtner, 2005). The coralecosystems in particular are an example of such syn-ergistic negative effects (Section 2.4; Hoegh-Guld-berg, 2005).

Acidification impacts on the food web are alsoconceivable. Different responses to increased CO2

concentrations, with respect to growth rates or repro-duction of an organism, could change the spatial aswell as temporal distributions of the species throughchanges in competition (Rost and Sültemeyer, 2003).Impacts that have already been observed in primaryproducers include a difference in the magnitude ofthe CO2 fertilization effect (which, for example,

70

Organisms Photosynthetic Crystal form of thecarbonate

Habitat

Coccolithophores yes Calcite Planktonic

Macroalgae* yes Aragonite or calcite

Benthic

Foraminifera nosome

CalciteCalcite

BenthicPlanktonic

Coralswarm-watercold-water

yes (in symbiosis)no

AragoniteAragonite

BenthicBenthic

Pteropods no Aragonite Planktonic

Non-pteropod molluscs* no Aragonite or calcite

Benthic or planktonic

Echinoderms no Mg-calcite Benthic

Crustaceans* no Calcite Benthic or planktonic

TTaabbllee 44..33--11Groups of calcifying marineorganisms. Calciumcarbonate occurs indifferent crystal forms.Aragonite dissolves morequickly than calcite at lowcarbonate ion concen-trations, but more slowlythan magnesium-rich calcite(Mg-calcite).* not all species of thisgroup are calcifiers.Source: after Raven et al.,2005

71Effects of acidification on marine ecosystems 4.3

favours coccolithophores over siliceous algae) andreduced calcification (which may be a disadvantagefor coccolithophores: Riebesell, 2004). In long-termstudies in the North Atlantic, it has been observedthat changes in the phytoplankton, due to the closecoupling with their predators, can be passed on firstto the algae-feeding zooplankton and then further to

the predatory zooplankton (Richardson and Schoe-man, 2004). A change in the species composition ofthe phytoplankton can thus impact on the zooplank-ton. In polar ecosystems it is conceivable thatreduced calcification by pteropods has effects on thehigher levels of the food web, although this is specu-lative and not easily predictable (Orr et al., 2005).

FFiigguurree 44..33--11Aragonite saturation andpresent occurrence of reeflocations for warm-watercorals (blue dots). (a) pre-industrial values (around1870, atmospheric CO2

concentration 280ppm), (b)present (around 2005,375ppm CO2), (c) future(around 2065, 517ppm CO2).The degree of aragonitesaturation (Ω) indicates therelative proportion betweenthe product of theconcentrations of calciumand carbonate ions and thesolubility product foraragonite. Locations with anaragonite saturation below3.5 are only marginallysuitable for reef-formingwarm-water corals, below 3they are not suitable.Source: Steffen et al., 2004


Conclusions about possible adaptive processes at theecosystem-structure level are also speculative, forexample, whether gaps that are created by acidifica-tion impacts can be filled by other species withoutsignificant effects upon overall productivity.

4.3.4Effects of acidification on fisheries

Acidification of the world’s oceans could also havean impact on fisheries. Direct toxic effects ofincreased atmospheric CO2 concentrations on fishare not expected because the threshold of acute sen-sitivity of fish to CO2 is beyond the predicted con-centrations (Pörtner, 2005; Section 4.3.1). When cal-cification is reduced, however, this can triggerchanges in the species composition of the phyto-plankton, and this, in turn, can have an impact all theway to the upper layers of the food web throughtrophic coupling (Richardson and Schoeman, 2004;Section 4.3.3). It cannot be ruled out that this kind ofchange in the structure and function of the marineecosystems can have an impact on the pelagic fish-eries, but with the present state of knowledge theprognosis remains very speculative (Raven et al.,2005).

Changes in growth and competitive conditions forthe species in tropical coral reefs will probably alsoaffect another important branch of fishery: millionsof people depend on subsistence fishery on coralreefs for their protein supply (Raven, et al., 2005),and the coral reefs themselves are threatened byacidification (Section 2.4). A large-scale loss of coralhabitats would doubtless have adverse effects uponthis fishery, with socioeconomic consequences thatare difficult to predict.

4.3.5Feedback of changes in calcification on the carboncycle

Overall, the ecological balances in the sea are shiftingto the detriment of calcifying organisms, and this mayaffect even the global biogeochemical cycles throughchanges in species compositions in marine phyto-plankton. The consequences of changing rates ofplankton calcification described here represent onlya small sample of all the interactions between the cli-mate system and the ocean, which are reviewed inSection 4.2.

The annual primary production in the ocean isapproximately 50Gt C, of which approximately 10Gtis exported to the deep sea by the biological pump.For this important process in the global carbon cycle,

which contributes to the sink function of the ocean, itmakes a great difference whether the production isby calcifying species like coccolithophores or by non-calcifying species, for example, siliceous algae.

Calcification by marine organisms always involvesCO2 production:

Ca2+ + 2 HCO3–➛ CaCO3 + CO2 + H2O

This carbonate ‘counter-pump’ becomes strongerwith increasing atmospheric CO2 concentration as aconsequence of the altered carbonate buffer capac-ity. Assuming constant calcification, this would causea future weakening of the sink effect of the sea. Butif biogenic carbonate formation is reduced as a resultof a pH decrease, then this effect can be overcom-pensated so that the sink effect may even bestrengthened. This would, however, only have aminor impact on the CO2 uptake by the ocean (Zon-dervan et al., 2001).A number of other effects furthercomplicate this picture (Riebesell, 2004): reducedcalcification could also reduce the density and thusthe sinking rates of particles to the deeper water lay-ers, slowing the carbon export by the biologicalpump. On the other hand, there is a possible acceler-ation of the sinking rates caused by the increased for-mation of extracellular polysaccharides (Engel et al.,2004). Present-day plankton blooms of cocco-lithophores cover large areas of the sea, up to hun-dreds of thousands of km2, and lighten the colour ofthe water because of their carbonate content. Theirabsence could therefore reduce the global albedo byup to 0.13 per cent, which would slightly accelerateglobal warming (Tyrell et al., 1999).The magnitude ofsome of these factors is not clear. The total effect ofall of these factors on the interactions betweenatmospheric CO2 concentration and marine biologi-cal production cannot currently be ascertained,which presents a need for increased research effortsin this area (Raven et al., 2005; IMBER, 2005).

4.4Guard rail: Ocean acidification

4.4.1Proposed guard rail

To prevent undesirable or high-risk changes to themarine food web due to aragonite undersaturation(Section 4.3), the pH value of near surface watersshould not drop more than 0.2 units below the pre-industrial average value of 8.18 in any larger oceanregion (nor in the global mean). A pH drop of 0.2units would correspond to an increase in the H+ ion

72

73Guard rail: Ocean acidification 4.4

concentration of around 60 per cent compared topre-industrial values. The decrease in pH so far of0.11 units since industrialization corresponds to arise of the H+ ion concentration of around 30 percent.The present average pH value of the ocean sur-face layer is 8.07 (Raven et al., 2005). Figure 4.4-1illustrates the WBGU acidification guard rail.

It is necessary, however, to further specify the spa-tial and temporal averaging to which the guard railrefers, because the pH value is subject to strong nat-ural variability. According to Haugan and Drange(1996), pH values vary up to 0.5 units worldwide,while local seasonal fluctuations can amount toaround 0.1 pH units (in high-production regionseven 0.2–0.3 units: Riebesell, personal communica-tion).

According to simulations by Caldeira and Wickett(2005), a stabilization of the atmospheric CO2 con-centration of 540ppm by the year 2100 would alreadylead to a pH decrease of the ocean surface layer of0.23 in the global average compared to the pre-indus-trial level, that is, at this CO2 concentration the acid-ification guard rail would already be overstepped. Astabilization at 450ppm by 2100 reduces the pH valueby 0.17, and so would presumably be consistent withthe acidification guard rail. It still needs to bereviewed, however, whether at this stabilizationvalue higher pH reductions could occur locally overlonger time periods, which, especially in the SouthernOcean, could lead to undersaturation of the surfacelayer with respect to aragonite. It should be notedthat this refers to the stabilization of CO2 itself andnot the stabilization level of greenhouse gases intotal, which is described by the CO2 equivalent.

4.4.2Rationale and feasibility

The largest threat to marine organisms due to acidi-fication is related to the solubility of calcium carbon-ate, which they need for the construction of theirshells and skeletal structure (Section 4.3). The moreeasily dissolved variant of calcium carbonate is arag-onite, which is used by corals and certain planktonspecies (Table 4.3-1). Calcifying marine organismsare important components of marine ecosystems, sotheir endangerment would represent a non-tolerableinterference with the Earth System.

If the concentration of carbonate ions falls belowthe critical value of 66µmol per kilogram, then theseawater is no longer saturated with respect to arag-onite, and marine organisms can no longer buildtheir aragonite shells.This needs to be avoided aboveall in the surface layer where primary productiontakes place. The danger of undersaturation for arag-onite is especially present in the Southern Ocean.According to Orr et al. (2005), simulations where thepH decrease averages about 0.25 already show aclear reduction of the vertical extent of the saturatedlayer, and undersaturation in some parts of theSouthern Ocean. It is the view of WBGU that such asituation should be avoided.

The pH value is a critical variable not only for cal-cification, but also for many other processes inmarine systems (for example, availability of nutri-ents). Within the past 23 million years the naturalfluctuations of the average pH value between glacialand interglacial periods lay within a range of slightlymore than 0.1 (Fig. 4.4-1), so that over a long timemarine organisms were able to adapt to a fairly nar-

FFiigguurree 44..44--11Variability of the averagepH value of the oceans inthe past and present, as wellas a projection for thefuture for an atmosphericCO2 concentration ofapprox. 750ppm. The redline indicates the WBGUguard rail.Source: after IMBER, 2005

Glacialat 190 ppm CO2

Preindustrial

Today

Future, at approx.750 ppm CO2

Anthropogeniceffect

Glacial-interglacialvariability

Present spatialand temporalvariability

WBGU Guard rail

8,5

8,4

8,3

8,1

8,2

8,0

7,9

7,8

7,71750 1800 1850 1900 1950 205020.000

b. p.2000 2100

Year

pH


row pH span that very rarely dropped below the min-imum in the surface-layer water (IMBER, 2005).Thisis a further argument for application of the precau-tionary principle, especially considering the gaps inthe scientific knowledge about the impacts of acidifi-cation (Section 4.3).

Because of the importance of the consequences ofocean acidification, research in this area should beintensified considerably (Section 4.6). As long asthere is no general scientific consensus about the tol-erable limit for the effects of acidification, a marginof safety according to the precautionary principleshould be observed.The suggestion of WBGU to pre-vent a pH decrease of more than 0.2 is orientedtoward the goal of avoiding an aragonite undersatu-ration in the ocean surface layer. If it is found thatother intolerable damages already occur beforereaching aragonite undersaturation, then the guardrail will have to be adjusted accordingly.

Because the CO2 input into the sea is caused by arise in atmospheric CO2 concentrations and there-fore by anthropogenic CO2 emissions, the pH drop inthe ocean can be limited by reducing emissions. Onceacidification has occurred, however, it is irreversible– as long as there is no possibility of lowering theatmospheric CO2 concentration the pH value of thesurface layer will not rise again in any foreseeablefuture. Overstepping the guard rail would thus beirreversible, which makes the precautionary principleparticularly relevant to this problem.

Compliance with the guard rail can be verifiedreliably by scientific means: for one, the pH value ofseawater can be determined directly, and for another,the average pH value of the sea can be derived frommeasurements of atmospheric CO2 concentrations.

The acidification and climate guard rails couldexhibit redundancies with regard to the measuresrequired to obey them, but they are not replaceableby one another: human-induced global warming iscaused by a group of greenhouse gases, with 60 percent of the effect from CO2. The only one of thisgroup responsible for ocean acidification, however, isCO2. Stabilization of CO2 at 450ppm by 2100 wouldreduce the pH value by 0.17, therefore staying withinthe allowable range of the acidification guard rail.Compliance with the global warming guard rail of2°C also requires a stabilization concentration of450ppm or less, depending on climate sensitivity.Therefore, observance of the climate guard railwould incorporate the acidification guard rail, underthe condition that CO2 is adequately taken intoaccount in the emissions reduction.

4.5Recommendations for action: Linking climateprotection with marine conservation

In the 1970s and 1980s, the phenomenon of ‘acid rain’became widely known. This problem is caused byemissions of acid-forming gases (mainly SO2 andNOX) from the combustion of fossil fuels. The issuewas taken up by the media under the catchphrase‘Waldsterben’ (forest dieback), and exerted consid-erable pressure upon policymakers and industry.Great technical and financial effort was subsequentlyinvested to fit large-scale power plants with flue gasscrubbers, mandate catalytic converters for cars, andembark upon broad-scale liming of forests and lakes.A comparison of the problem of acid rain with thealready advancing acidification of the oceans showsthe latter to be an issue significantly more serious.The media and policymakers, however, are express-ing negligible interest compared to acid rain. Indeed,the problem is being practically ignored. Policymak-ers are therefore called upon to recognize the fullimpact of ocean acidification and to take measures ofa scope and effectiveness comparable to thoseadopted to tackle acid rain.

4.5.1Reappraising the role of CO2 in climate protectionpolicy

The release of CO2 has particularly far-reaching con-sequences for marine ecosystems. Firstly, CO2 acts asa greenhouse gas, altering the radiation balance ofthe atmosphere and thus contributing to globalatmospheric warming and, as a further consequence,to the warming of the oceans. Secondly, a large pro-portion of the CO2 emitted by human activities dis-solves in seawater, where, in addition to the warming,it causes chemical changes. In view of these particu-larly harmful effects of CO2 upon the oceans, it isessential that climate policy give special attention tothis specific greenhouse gas.

Need for actionCompliance with the WBGU guard rail for oceanacidification will only be possible if the increase ofthe atmospheric CO2 concentration is limited. Engi-neering approaches, such as liming the surface layersof the oceans, are unrealistic considering the scale ofthe problem (Raven et al., 2005). However, over atime scale of centuries, the acidified surface waterwill mix down into the deep sea through ocean cur-rents. One option for action is to stabilize the atmos-

74

75Recommendations for action: Linking climate protection with marine conservation 4.5

pheric CO2 concentration; this would lead over thelong term to an acidification of deeper waters of theoceans until the pH of the surface layers is reached.An alternative option would be to agree a maximumlimit on the total amount of CO2 emitted to theatmosphere by human activities; that approach couldcause the atmospheric CO2 concentration to dropagain over the medium term and could prevent theacidification of the deep sea.

Legal settingThe present climate policy instruments do not takeinto account the aspect of ocean acidification causedby CO2 input. WBGU takes the view that the UnitedNations Framework Convention on Climate Change(UNFCCC) does indeed establish an obligation totake into account the impacts of climate change uponthe oceans, regardless of the circumstance that thisaspect was not a priority when the UNFCCC wasconcluded, and was not covered by the Kyoto Proto-col when reduction commitments were set.The ratio-nale is as follows: Under Art. 1 para. 3 UNFCCC,‘cli-mate system’ means the totality of the atmosphere,hydrosphere, biosphere and geosphere and theirinteractions.The term ‘climate system’ is thus definedin such a comprehensive way that it includes theoceans, which are a part of the hydrosphere, as well asthe interactions of the oceans with the atmosphereand the biosphere. The UNFCCC objective estab-lished in Art. 2, ‘stabilization of greenhouse gas con-centrations in the atmosphere at a level that wouldprevent dangerous anthropogenic interference withthe climate system’, thus also covers the impacts ofincreasing greenhouse gas levels upon the oceans.With respect to acidification, the meaning of Art. 2UNFCCC can be concretized as follows: CO2 is agreenhouse gas, and an excessive CO2 concentrationin the atmosphere leads to dangerous anthropogenicinterference with marine ecosystems, for CO2 dis-solves in water and causes acidification (Sections 4.1and 4.3). The oceans are a part of the hydrosphere,and marine organisms are a component of the bios-phere. The problem of acidification is thus one ofinteraction among the atmosphere, hydrosphere andbiosphere, all of which are components of the climatesystem (Art. 1 para. 3 UNFCCC). There can thus beno doubt that the objectives of the conventioninclude preventing a dangerous acidification of theoceans. Furthermore,Art. 2 UNFCCC also states thatecosystems should be able to adapt naturally to cli-mate change. The speed of acidification observedtoday calls compliance with this requirement intoquestion: for instance, the adaptive capacity ofmarine ecosystems can be overstretched if the arago-nite saturation horizons in the Southern Ocean riseto the surface (Section 4.3).This presents an immedi-

ate need to limit acidification and adopt appropriatemeasures under the UNFCCC.

RecommendationsWBGU argues against this backdrop that climatepolicy needs to take all impacts upon the marinehabitat into account. In the negotiations on the sec-ond commitment period under the Kyoto Protocolnow commencing, the German federal governmentshould work to ensure that the direct adverse effectsof CO2 emissions upon the oceans are taken intoaccount. The desired stabilization of atmosphericgreenhouse gas concentrations should be set in sucha way that ocean acidification is adequately limited.This implies that CO2 should not be viewed only aspart of a basket of various greenhouse gases. Theatmospheric CO2 concentration rather needs to bestabilized specifically, regardless of the reduction ofother greenhouse gases – at a level permitting com-pliance with the WBGU acidification guard rail (Sec-tion 4.4).

To achieve this goal, it may be necessary to definea CO2 emissions ceiling for individual states orgroups of states in addition to existing reductioncommitments. This CO2 cap would then need to beobserved as a complement to the other commit-ments. The precise effects of this and any furtherpotential instruments still need to be clarified.A par-ticularly important aspect in this regard is the possi-ble need for adaptation of the existing flexible mech-anisms (emissions trading, Clean DevelopmentMechanism and Joint Implementation).

It would not, however, be necessary to define aseparate ceiling for CO2 if, firstly, the internationalcommunity were to agree to reduce greenhouse gasemissions to a level ensuring compliance with theWBGU guard rail on climate protection, and, sec-ondly, the relative proportion of CO2 within overallgreenhouse gas emissions does not change signifi-cantly. The CO2 reduction needed in this case wouldmost probably suffice to prevent transgression of theacidification guard rail.

4.5.2Taking shipping sector emissions into account

As a part of efforts to stabilize the atmospheric CO2

concentration, the CO2 emissions generated by oceanshipping and international aviation should be inte-grated more closely into emissions reduction strate-gies. No quantitative reduction commitments haveyet been agreed for either sector. WBGU recom-mends closing these regulatory gaps by integratingthe CO2 emissions generated by international ship-ping and aviation into negotiations on future reduc-

tion commitments within the Kyoto process. Presentestimates suggest that worldwide CO2 emissionsfrom shipping amount to about 2 per cent of globalemissions. Over the past decade, the rate of increasein shipping emissions was more than twice that fortotal emissions (Bode et al., 2002; IEA, 2002). Thisillustrates the urgent need for action.

Besides emitting CO2, ocean shipping also gener-ates pressures upon marine and coastal ecosystemsby emitting pollutants, nutrients and sediment parti-cles. Controls on ocean shipping thus present a start-ing point for linking climate protection with marineconservation at the level of legal instruments.

In view of the relatively good environmental per-formance and significant economic importance ofocean shipping, regulatory controls need not aim toreduce the volume of shipping traffic. The goal israther to create incentives for technological innova-tions and improvements in environmental manage-ment that contribute both to abating ocean pollutionand preventing atmospheric CO2 emissions. As ameans to this end, WBGU recommends levyingcharges on the use of the oceans by shipping(WBGU, 2002).

This instrument highlights the connectionbetween the utilization of the environmentalresources represented by ‘the oceans’ and ‘theatmosphere’ and the utilization-related impairmentof these resources. A charge signals the scarcity ofenvironmental resources and the cost of their provi-sion. The economic players burdened by a chargereceive an incentive to modify their utilization ofglobal environmental goods and to make it more sus-tainable (WBGU, 2002).

WBGU has set out the options for designing auser charge system in detail elsewhere (WBGU,2002). A proposed user charge regime applied to theEuropean Union area alone could generate annualrevenues of €400–700 million. WBGU proposes thatthe financial resources thus received be earmarkedfor marine conservation purposes, in order to createa substantive link between the generation and reduc-tion of pressures upon the oceans (WBGU, 2002).


Acidification and marine ecosystemsThe physiological effects of acidification on marineorganisms, especially on calcifying ones, and theimpacts on the marine ecosystem are insufficientlyunderstood. Physiological experiments are neededwith moderately increased CO2 concentrations, asare experiments exploring the effects on marine foodwebs (trophic coupling among phytoplankton, zoo-

plankton and fish), and studies of possible physiolog-ical adaptation processes on an evolutionary basis.

Biogenic calcification and the carbon cycleOur understanding of the interactions between calci-fying plankton, the biological pump, and the globalcarbon cycle shows similar gaps, so modelling of thenet effect is not yet possible. Modelling studies there-fore need to be carried out on the acidification-related reduction of biological export productiondue to decreased mineral ballast (carbonate shells).

Further impacts of climate change Acidification is probably just one of many changesthat will take place in the biogeochemistry of theoceans due to anthropogenic greenhouse gas emis-sions, or through climate change. Other aspects, suchas the effects on oxygen balance and nutrient supplyin the sea, are poorly understood and urgently needfurther study in order to recognize critical develop-ments in good time.

Future CO2 uptake by the oceanCO2 uptake by the ocean plays a key role in climatechange. Interactions among the atmospheric radia-tion balance, the chemical composition of the atmos-phere and the physical, chemical, and biologicalchanges in the ocean should therefore receiveincreasing attention.

International research programmesPromotion of projects by international research pro-grammes (e.g., SOLAS, 2004; IMBER, 2005) thataddress the questions above is recommended.

CO2 and climate protectionIn the event that a specific reduction commitmentfor CO2 proves to be necessary, the potential ways ofdesigning such a commitment need to be developedand evaluated. In addition, the implications forKyoto mechanisms (especially CDM and emissionstrading) need investigation.

4 Ocean acidification76

CO2 storage in the ocean and under the sea floor

Great and growing hopes have been pinned of lateupon the sequestration of CO2 as a means of climatemitigation (IEA, 2004). IPCC discussed this theme indepth in a recent Special Report (IPCC, 2005). Esti-mates expect carbon dioxide capture and storage(CCS) to be market-ready by 2015 (IEA, 2004).Within 50 years, 20–40 per cent of the CO2 emissionsarising from the combustion of fossil fuels could beseparated, captured and stored (IPCC, 2005), pro-vided that research and development intensify signif-icantly (IEA, 2004). Sequestration technology hasdirect relevance to the present report, as it alsoincludes the storage of CO2 in the ocean and underthe sea floor (Box 5.3-1).

5.1CO2 sequestration

5.1.1Potential and costs

The technology of carbon dioxide sequestration hasthree components: CO2 capture, transport and stor-age (IEA, 2004). Storage locations under considera-tion include sub-seabed geological formations, thewater column of the ocean, and onshore geologicalformations such as depleted oil and gas fields andunminable coal seams. Chemical fixation to metaloxides is conceivable, although this process is cur-rently regarded as unsuitable in view of the enor-mous energy consumption and very high costs asso-ciated with it (IPCC, 2005).

The storage capacity of depleted oil and gas fieldsis approximately 30 to 40 times the current annualCO2 emissions from the combustion of fossil energycarriers.The storage potential through Enhanced OilRecovery (EOR), whereby CO2 is injected into cavi-ties in order to increase oil yield, is estimated as 3 to5 times the annual CO2 emissions. Estimates forabsorption in coal seams vary between 13 per centand nine times annual CO2 emissions. Saline aquifersunder the sea may be able to hold 40 times the annual

CO2 emissions or more (IPCC, 2005). However, withthe exception of EOR, little practical experiencerelating to geological storage is available, and thesuitability of potential reservoirs is not clear.

Large point sources such as large fossil powerplants near potential storage locations are regardedas particularly attractive for CCS.Typically, 80–90 percent of the CO2 generated in fossil power plantscould be captured. However, the process requiresenergy, resulting in an increase in fuel consumptionby 16–31 per cent (or even 70 per cent if the technol-ogy is retrofitted to existing lignite-fired powerplants). Transportation and injection of CO2 requirecomparatively small amounts of energy. Comparedto the amount of emissions avoided, about 20–40 percent more CO2 has to be put into storage – and morethan twice the amount if existing lignite-fired powerplants are retrofitted.

CO2 emissions from large-scale biomass facilitieswould also be suitable for sequestration. This wouldcreate an actual CO2 sink, since the carbon containedin the biomass was previously removed from theatmosphere via photosynthesis.

The costs of CO2 capture are currently estimatedat US$11–57 per t of CO2, depending on the fuel, theage and type of the power plant, and the capturetechnology used (IPCC, 2005). Pipelines are state ofthe art for CO2 transportation. In the USA alone,40Mt CO2 are transported each year via pipelineswith an overall length of 2500km. However, for largedistances transport by ship is more economic thanpipelines. The costs for transporting 1 tonne of CO2

by ship are approximately US$15–25 per 5000km,compared with US$4–30 per 1000km via pipelines(IEA, 2004; IPCC, 2005). The costs for injection andstorage are comparatively low, estimated atUS$0.5–8 per t of CO2. In addition, there are minorcosts for monitoring and maintenance of the reser-voirs. The total costs of sequestration involving stor-age in the ocean or under the sea floor thereforerange between US$20 and 100 per t of CO2.

Based on current knowledge, sequestration of theCO2 released during power generation would lead toincreases in generating costs per MWh amounting to

5

5 CO2 storage in the ocean and under the sea floor

US$12–34 in new power plants. For retrofitted lig-nite-fired power stations the cost increase is esti-mated at US$33–44 per MWh (IPCC, 2005). Currentgenerating costs are around US$25–55 per MWh,depending mostly on fuel prices, which means thattotal generating costs including sequestration wouldbe US$45–80 per MWh. This range is comparablewith many wind and small-scale hydroelectric plants(Box 5.3-2). Sequestration would increase powergeneration costs in fossil power plants by 30–60 percent for new plants. Retrofitting existing plants maytriple costs. Optimistic forecasts assume that seques-tration costs are likely to come down significantly by2030. However, based on renewable electricity gen-erating costs of US$10–20 per MWh (IEA, 2004) andexpected increases in fossil fuel prices in the longterm, electricity generation from renewables is likelyto become an increasingly cost-effective option.

5.1.2Risks and sustainability

The uncertainty regarding the environmental sus-tainability of sequestration is more significant thanthe uncertainties relating to cost development.A dis-tinction has to be made between three types of risk.1. Risk of accidents: Similar to natural gas pipelines,

CO2 pipelines may be affected by leakage. CO2

concentrations of more than 7–10 per cent in airendanger health and life. However, experiencewith existing pipeline systems shows that majordamage to pipelines is very rare. In addition, therisk can be reduced further through improvedpipeline design and monitoring. Sudden escape oflarge quantities of CO2 is also conceivable duringCO2 injection into the repository. In addition, sim-ilar to EOR or natural gas storage, stored CO2

may escape abruptly, e.g. due to inadequate seal-ing of the repository (IPCC, 2005). However, thistype of major accident is regarded as unlikely inconjunction with CO2 storage. The immediateimpacts of such an incident would be significantlylower at sea than in inhabited areas, where severe,in extreme cases fatal impact on humans wouldhave to be expected.

2. Potential impact on marine ecology: This is mainlyassociated with CO2 disposal in seawater, whichWBGU regards as unacceptable. The issue is dis-cussed in Section 5.2.

3. Continuous slow escape of stored CO2: This risk ishighly significant in the context of long-term cli-mate change mitigation. While the IPCC SpecialReport (IPCC, 2005) contains no specific data onacceptable leakage rates, a simple rough calcula-tion can provide some guidance. The cumulative

emissions in the different SRES scenarios for1990–2100 vary between 1000Gt C (B1 scenario)and 2200Gt C (A1FI scenario) (IPCC, 2000). Inorder to comply with the 2ºC climate guard rail,the cumulative emissions to the atmosphere fromthe present need to be limited to 500Gt C (Meins-hausen, 2006). Compared with a medium-levelscenario assuming emission of 1500Gt C by 2100,around 1000Gt C would have to be mitigated. Ifthis quantity were to be sequestered, with a leak-age rate of 0.1 per cent per year (i.e. a retentionperiod of 1000 years) 1Gt C would escape uncon-trolled every year. However, in order to complywith the 2ºC guard rail, a maximum of 1Gt C oftotal emissions per year would be acceptable inthe long term (from about 2200), even for the caseassuming an average climate sensitivity of 3ºC(Caldeira et al., 2003). Thus even assuming amedium-level emissions scenario, which does notrepresent the worst case, leakage from CO2 stor-age sites alone would represent 100 per cent ofadmissible CO2 emissions in the long term.The sit-uation is even more problematic if less optimisticassumptions are made: Climate sensitivity mayprove to be higher, other greenhouse gases (e.g.methane, see Chapter 6) may contribute to warm-ing more strongly than assumed, or the proposed2ºC guard rail may prove to be too high in the longterm, e.g. in the event that it triggers the melting ofGreenland ice (see Chapter 3). Overall, no morethan one-tenth of the above-mentioned leakagerate would therefore appear to be acceptable, i.e.0.01 per cent per year, corresponding to a reten-tion period of 10,000 years. Therefore, sequestra-tion can only be regarded as an acceptable climatemitigation technology if long-term CO2 storagefor at least 10,000 years can be guaranteed.

5.2Ocean storage

Two basic options are under consideration for carbonsequestration in the ocean: physical-chemical disso-lution in the seawater and, in the broadest sense, bio-logical-engineered storage in marine ecosystems, pri-marily through iron fertilization. In the following,only the physical-chemical techniques will be dis-cussed in detail.This report does not explore the con-cept of using permanent input of iron to trigger algalblooms and thereby increase the sink potential of theocean in marine areas where the micro-nutrient ironis the limiting factor for primary production (notablythe Southern Ocean). The expected quantitativeeffect is fairly low (as a comparison with palaeocli-matological data leads one to presume), and there is

78

79Ocean storage 5.2

doubt that the permanence of storage is sufficient(Section 5.1.2). Furthermore, the risks of large-scaleiron fertilization in terms of indirect effects on themarine ecosystem are hard to estimate. WBGU hasalready explained elsewhere the reasons for its rejec-tion of iron fertilization of the ocean (WBGU, 2004).

5.2.1Storage and residence time of CO2

Direct injection into seawater is one form of CO2

storage that is under discussion. The CO2 content ofthe sea surface equilibrates relatively quickly withthe atmosphere, so that an artificial increase of CO2

in the surface water would result in outgassing to theatmosphere within a short time. Introduction into thedeep sea could, in contrast, ensure a longer residencetime of carbon in the sea. The CO2 injected therecould remain isolated from the atmosphere for sev-eral centuries (IPCC, 2005), but over longer timeperiods the equilibrium between atmospheric CO2

concentration and that in the sea would be re-estab-lished.Then, depending on the atmospheric CO2 con-centration, between 65 and 80 per cent of anthro-pogenic CO2 would be stored in the sea, regardless ofwhether the CO2 has been emitted to the atmosphereor injected into the ocean (Caldeira et al., 2005). Theinjection of CO2 into seawater could thus reduce apeak concentration of CO2 in the atmosphere, but ithas no influence on the long-term stabilization levelof atmospheric CO2. Thus, independent of the conse-quences for the marine ecology (Section 5.2.2), itdoes not represent a sustainable solution for theproblem because future generations would be bur-dened with irreversible effects.

Another technological option would be the stor-age of CO2 as a liquid or hydrate on the sea floor,which would only be possible in water depths below3000m due to its greater density there. Without aphysical barrier, however, the CO2 would slowly dis-solve from such reservoirs into the overlying watercolumn. So this technology would also only lead to apostponement of the consequences of climatechange, but not to their mitigation. None of the tech-nological possibilities being discussed for storage inseawater have been tested in field studies at a mean-ingful scale. Approval has not been given for any ofthe research projects so far proposed, not even forinjecting just a few tonnes of carbon dioxide into thedeep sea.

5.2.2Impacts of CO2 storage on deep-sea organisms

Just as in the surface layer, the direct injection of CO2

into the deep sea also changes the chemical and phys-ical characteristics of the seawater. Initially thisaffects the direct surroundings of the location ofintroduction, for example, the end of the pipelinethrough which the liquid CO2 flows into the deep sea.Here, as simulations indicate, dramatic changes in thelocal pH values of up to several units can occur.Through technical solutions that lead to faster dilu-tion (such as a pipeline towed by a ship), the maxi-mum local pH change can be reduced. In the some-what broader surroundings (several kilometres), therate of dilution is essentially determined by oceancurrents, so that the chemical and physical impactscan be estimated with ocean circulation models. Forexample, with an input of 0.1Gt C per year (which isless than 2 per cent of the industrial emissions andaround 5 per cent of the present CO2 input throughthe sea surface caused by anthropogenic CO2 levelrise in the atmosphere), in up to 0.01 per cent of theocean volume the pH value could drop by 0.3 unitsover a period of 100 years (Caldeira et al., 2005). CO2

storage in the deep sea could thus have seriousimpacts on the deep-sea ecosystem. Deep-sea organ-isms develop very slowly, their metabolic rates arelower and life expectancy is greater than of organ-isms in other ocean layers (IPCC, 2005). During theirevolution, the inhabitants of the deep-sea ecosystemhave adapted to special living conditions, with typi-cally very stable temperatures and pressures, and rel-atively constant CO2 concentrations (except at vol-canic CO2 vents). Such constant environmental vari-ables do not demand rapid adaptive strategies. Thus,it has to be expected for the possible storage of CO2

on the sea floor, as well as for leakage of a storagereservoir below the sea floor, that the ecosystemaffected will be critically damaged, or will take a longtime to recover from a change in the environment(IPCC, 2005).

Very little is known about the organisms in thedeep sea in general, their life forms and interactions.So far, the direct effect of CO2 on marine organismshas mainly been investigated in the laboratory. Stud-ies involving field observations are greatly lacking,except for a few experiments with small CO2 plumeson the sea floor and investigations of volcanic CO2

vents (Pörtner, 2005).In one of these in-situ experiments off the coast of

California, liquid CO2 was injected at 3600m in orderto study the survival and behaviour of the deep-seafauna after direct contact with CO2 (Barry et al.,2004). Depending on pH changes and distance from


the CO2 plume, the survival rate of the animals var-ied. Flagellates, amoebas and nematodes in the sedi-ment zone near the CO2 source showed a high mor-tality. In another study, the scents of prey animalswere combined with the extrusion of CO2 (Tamburriet al., 2000). Fish and invertebrates were attracted bythe scents and appeared to some extent to remainrelatively undamaged, even at a distance of just a fewcentimetres from the CO2 source, in spite of the lowpH value. Carrion-eating hagfish, attracted by thescent of the prey, did not seek to escape narcotizationunder the high CO2 content. Tyler (2003) thereforefears that animals that die through contact with CO2

introduced into the deep sea could attract larger car-rion eaters, who would then likewise be killed by theCO2 plume. Squid and other invertebrates may reactmore sensitively to high CO2 concentrations thanvertebrates (Pörtner et al., 2004) because their bodyfluids contain no haemoglobin, which helps protectthe body from large pH fluctuations. So even a small,local CO2 plume could have wide-reaching effects onits surroundings.

Risks also arise from outgassing into the atmos-phere. Two catastrophes occurred in the 1980s whenlarge CO2 plumes from gas-saturated deep waterescaped into the atmosphere from the volcanic LakesMonoun and Nyos in Cameroon. The disaster atLake Nyos had devastating consequences: around 80million m3 of CO2 were expelled, taking the lives of atleast 1700 people and several thousand animals up toa distance of 10km from the lake (Kling et al., 1987;Clarke, 2001). There is sparse information in the lit-erature on whether Lake Nyos harboured life of anykind before the catastrophe, and how the gas plumeaffected this biotope. Freeth (1987) has reportedthat, in spite of otherwise favourable living condi-tions, the local population had neither seen fish in thelake before the catastrophe, nor were fish cadaversfound after the event.

If a large plume of CO2 pumped into the seashould rise to the sea surface or into higher water lay-ers, the possible ecological results can only be specu-lated. In summary, the largely incalculable ecologicalrisks also support a general prohibition of CO2 stor-age in seawater.

5.2.3Present international law

The relevant body of international law relating toCO2 storage in the ocean and below the sea floor canbe summarized as follows: according to the Conven-tion on the Prevention of Marine Pollution by Dump-ing of Wastes and Other Matter – the London Con-vention of 1972 – the disposal of certain wastes and

other matter (listed in Annex I to the convention)into the sea is forbidden. Further wastes and matterlisted in Annex II to the convention may only be dis-posed of with prior special permission. Other wastesand matter may be disposed of under a prior ‘gen-eral’ permit. Since 1 January 1996, the ‘black list’ ofAnnex I includes industrial waste (No. 11), whichmeans ‘waste materials generated by manufacturingor processing operations’. It can be assumed that sep-arated CO2 is derived from such operations and istherefore industrial waste within the meaning ofAnnex I. However, with respect to matter whose dis-charge into the ocean is prohibited, the conventioncontains an important exception in connection withthe extraction of mineral resources: according to Art.III, para. 1(c) of the London Convention, the ‘dis-posal of wastes or other matter directly arising from,or related to the exploration, exploitation and associ-ated offshore processing of seabed mineralresources’ is not covered by the provisions of theconvention. In other words, the disposal of CO2 thatis generated by the production of oil or natural gas atsea is permitted under the Convention, as long as thecorresponding processing operations are carried outat sea.

Basically the same legal position exists under theProtocol of 1996, although the approach is different:the Protocol, which will replace the Convention inthe future but has not yet been ratified by a sufficientnumber of signatories and is therefore not yet inforce, contains a general prohibition of discharge intothe sea, combined with a list (Annex 1) of exceptions.CO2 is not included among these exceptions. Thismeans that the discharge of CO2 would be essentiallyprohibited under the Protocol once it enters intoforce. But, according to the Protocol, the dischargewould still be allowed when the CO2 is derived fromthe recovery of oil or natural gas at sea and the pro-cessing also takes place there (Art. 1, para. 4.3).

5.3Sub-seabed geological storage

5.3.1CO2 injection into the geological sub-seabed

Injecting CO2 into geological formations below thesea floor is basically no different than the procedureon land. Saline aquifers, for example, also providerepositories, and pressurized injection of CO2 into oilformations could facilitate the extraction of oil. Thetechnical systems just have to be adapted for theexisting conditions.The appropriate monitoring tech-niques, however, are very different on land and in the

80

81Sub-seabed geological storage 5.3

sea. There are also some differences with respect tosafety technology (Section 5.3.3.4).

Not only are great research efforts presently beingcarried out on CO2 storage in the seabed (CSLF,2005), but practical experience already exists in thisfield, and further projects are planned (BellonaFoundation, 2005; Deutsche BP, 2005).When chargeson CO2 or the prices for emission rights rise, seques-tration becomes more economically attractive, andcompanies can be expected to apply increasingefforts in addition to the Sleipner project (Box 5.3-1)and EOR (Section 5.1). The Norwegian companyStatoil is already considering the transport of ‘for-eign’ CO2 through pipelines to the company’s Sleip-ner gas platform, and storing it there in the CO2 for-mations already in use under the sea.

5.3.2Risks and sustainability of CO2 storage in theseabed

Various scenarios are imaginable for the escape ofCO2 from formations under the sea floor. If the CO2

emerges at a depth where it occurs as hydrate, thenthe least damage can be expected. But when the CO2

dissolves in water it contributes to acidification of thesea. The conceivable harmful consequences of leaksfor marine organisms have already been described inSection 5.2.2. In cases of very large-volume leaks, theCO2 could also reach the surface, which would, forone, contribute to the enrichment of CO2 in theatmosphere and, for another, present a health risk inthe immediate surroundings. But as long as the stor-age site is not directly on the coast near human set-tlements, the human health risk is significantly lower

BBooxx 55..33--11

TThhee SSlleeiippnneerr pprroojjeecctt

The Sleipner Platform in the North Sea is located approxi-mately 250km from the coast of Norway. It is the first com-mercial project for CO2 storage in a saline aquifer under thesea floor. CO2 separated from natural gas here is trans-ported locally to a depth 800m below the sea floor.The stor-age of CO2 here is economically interesting because the sep-aration of CO2 from the gas is necessary in any case for latertechnical use, and the Norwegian government would tax itsemission into the atmosphere. Since October 1996 around1Mt CO2 has been injected annually into the sub-seabed. Bythe beginning of 2005 more than 7Mt CO2 had been injectedinto the aquifer. By the end of the project the total shouldbe around 20Mt CO2. The formation has a total capacity of1–10Gt CO2.

The project is being observed scientifically, in part toinvestigate safety and permanence of storage. Initialresearch results indicate that a dense cap rock seals the for-mation at the top, preventing the leakage of CO2. Simula-tion calculations covering hundreds of thousands of yearssuggest that the CO2 will dissolve in the pore waters andthen sink to the bottom of the formation.The probability oflong-term leakage is minimal, so that the gas, according tothese calculations, should not escape into the North Sea forthe next 100,000 years. Even after a million years, only a mil-lionth of the CO2 should escape. This storage could there-fore fulfil the required holding time of more than 10,000years (Section 5.1.2), but these conclusions will have to bebetter documented scientifically.

Sources; IPCC, 2005; Statoil, 2005

FFiigguurree 55..33--11The Sleipner project in theNorth Sea, simplifiedrepresentation. The gasproduction comes from theSleipner East gas field. Thecaptured CO2 is injectedinto the Utsira sandstoneformation. The small pictureshows the position andextent of the Utsiraformation in the North Sea.Source: Statoil, 2005


than for storage on land. Even where people are inthe vicinity, the probability of dangerously high CO2

concentrations in the ocean environment isextremely low because, in contrast to the situation onland, CO2 lakes cannot form. As a rule, such CO2

lakes can only form and persist in depressions onland that have no or poor drainage.

As discussed in Section 5.1.2, a retention time forCO2 of at least 10,000 years is required for long-termsustainability.

5.3.3Regulating sub-seabed geological storage

Considering that global CO2 emissions are rising, theoption of storing CO2 in geological formations deepbelow the sea floor should not be dismissed com-pletely. However, such sub-seabed geological storageis not altogether unproblematic (Section 5.3.2). Forone thing, a release of CO2 to the atmosphere cannotbe excluded entirely. This can be caused by technicalfaults or by accidents arising in the transport, injec-tion and storage process. It may also be due to theselection of inappropriate geological formations.Current knowledge indicates that, under certain geo-logical and technological preconditions, leakagerates may be acceptable (<0.01 per cent per year).There is a need for substantial further research, how-ever, to be able to verify this with sufficient certainty.Issues in particular need of clarification include thecriteria that geological formations must meet, andhow any escape of the gas to seawater could be mon-itored and quantified.

Moreover, an all too strong political and economicfocus on the sequestration option might causeneglect of far superior climate mitigation strategies,such as improving energy efficiency and switching torenewable energies. To attain the goal of sustainableenergy systems, it is these superior options that par-ticularly require political support, innovation and theemployment of scarce resources (WBGU, 2004). Ahigh renewable energy potential is available in theocean and above the sea surface (Box 5.3-2).

WBGU therefore views sub-seabed storage ofCO2 as being, at most, a transitional option comple-menting other options (WBGU, 2004). Its deploy-ment should be limited and regulated (Section5.3.3.4).

5.3.3.1Provisions under the international law of the sea

The 1972 London Convention and its 1996 LondonProtocol permit the storage of CO2 in sub-seabedgeological formations if the sequestered CO2 origi-nates in the course of processing the mineralresources of the seabed (Section 5.2.3; the sameapplies to placement of CO2 in seawater). This is thecase, for example, with the Sleipner project (Box 5.3-1).

In contrast, it has not yet been clarified unequivo-cally whether the 1972 London Convention or, infuture, the 1996 London Protocol permits sub-seabedstorage, for instance in saline aquifers, of CO2 thatwas separated on land (IEA, 2005). Article III, para.3, of the London Convention defines ‘sea’ as ‘allmarine waters’. There is some controversy as towhether this definition means that the seabed andthe subsoil thereof fall within the scope of the con-vention. In response to a survey conducted by IMO,Germany argued in favour of construing the term ‘allmarine waters’ to include the seabed and the subsoilthereof, as this would be in line with the history andpurpose of the convention.The 1996 Protocol definesin Art. 1, para. 7, the term ‘sea’ more precisely, namelyas ‘all marine waters other than the internal waters ofStates, as well as the seabed and the subsoil thereof;it does not include sub-seabed repositories accessedonly from land’. This definition, however, has alsogiven rise to controversy over the depth to which thesubsoil reaches. In the above-mentioned IMO survey,Germany argued in favour of construing the term ascomprehensively as possible, too.

When construing the treaty wording, however, itneeds to be taken into account that the issue of CO2

sequestration, including CO2 storage in the ocean orunder the sea floor, was not on the agenda when the1972 London Convention was negotiated, nor whenits 1996 Protocol was elaborated. It is therefore notpossible to draw conclusions from the wording of thetreaty about the will of the participating states withrespect to how to handle CO2.The parties to the Lon-don Convention are now addressing this issue inten-sively (IMO, 2004), for instance at the 27th Consulta-tive Meeting of the parties held in October 2005. Inview of the numerous gaps in knowledge and theunresolved issue of whether placement of CO2 in theseabed should be covered by the London Conven-tion and/or the London Protocol, that meetingagreed to debate the issue in greater depth at the28th Meeting. If the parties resolve to permit theplacement in the seabed of CO2 sequestered fromseparation processes on land,Annex 1 to the LondonProtocol may need to be amended; this would also beexpedient in order to provide clarification. The pre-

82

83Sub-seabed geological storage 5.3

sent state of knowledge thus indicates that it wouldbe necessary to take account of Art. 31, para. 1, of theVienna Convention on the Law of Treaties, accordingto which a treaty shall be interpreted in good faith inaccordance with the ordinary meaning to be given to

the terms of the treaty in their context and in the lightof its object and purpose.

BBooxx 55..33--22

MMaarriinnee rreenneewwaabblleess

In addition to their role within the climate system, theoceans offer options for active mitigation of anthropogenicclimate change. On the one hand, increased utilization ofrenewables from the sea can substitute fossil energy carriersand therefore reduce associated CO2 emissions. On theother hand, CO2 storage in suitable geological formations inthe seabed may offer an additional man-made sink for thisgreenhouse gas. The potential for marine renewables isbriefly outlined below, followed by a rough comparison ofthe respective costs for the two options.

Potential for marine renewables Commensurate with their proportion of the Earth’s surface,the oceans receive more than 70 per cent of the solar inso-lation and almost 90 per cent of the wind energy (Czisch,2005).They therefore hold the majority of the global renew-ables resources. However, from today’s perspective onlyfractions of this theoretically available energy is technicallyand cost-effectively usable. In addition, the potential isreduced by a wide range of competing uses, particularlyalong densely populated coastlines. The sustainable poten-tial is reduced further by the fact that environmentalaspects have to be taken into account (WBGU, 2004). Forexample, any expansion of renewables must comply withthe ecosystem guard rail (20–30 per cent of marine ecosys-tems designated as protected areas; Section 2.5). The over-all area available for sustainable utilization of renewableenergy is therefore reduced significantly.• Wind energy: Studies on the European offshore wind

energy potential (Sea Wind Europe, 2003) assume aninstalled capacity of 111GW and associated annual gen-eration of 340TWh by 2015, equivalent to approximately10 per cent of the technical potential. Siegfriedsen et al.(2003) conclude that outside the European Unionapproximately 4600TWh per year could be generated byoffshore wind energy converters. With a total of5000TWh per year, one-third of current global electric-ity demand of approx. 15,500TWh per year would becovered by offshore wind energy. In 19 of 20 countrieswith the largest potential outside the EU, more than 10per cent of electricity demand could be covered by off-shore wind energy converters by 2020. Of the differentenergy forms examined, wind energy is the most signifi-cant one in terms of potential and implementation.

• Wave energy: Wavenet (2003) estimates the global tech-nological potential as 11,400TWh per year. The globalsustainable generating potential of wave energy isapprox. 1700TWh per year, i.e. more then 10 per cent ofcurrent global electricity demand.An annual generationfigure of 9TWh is assumed for the EU by 2020. Waveenergy is not expected to make a significant contribution

to global electricity demand until some time in thefuture.

• Tidal energy: Strong sea currents occur near coaststhrough tidal and other effects. The potential for energygeneration from such currents in North America,Europe, South-East Asia and Australia is estimated at120TWh per year. The total global sustainable potentialis likely to be several hundred TWh per year. In 5–10years’ time, sea current turbines could experience a sim-ilarly dynamic development as that currently seen withoffshore wind energy converters.

• Energy from osmosis: A further energy production tech-nique is based on the utilization of osmotic pressurebetween freshwater and seawater (e.g. in estuaries)using special membranes with high salt retention. Thistechnology is currently only available at the laboratoryscale. Globally, a total of 730GW could be achievablefrom rivers with flows of more than 500m3 per second.The sustainable potential, taking into account ecologicalguard rails and shipping-sector requirements, is esti-mated to be around 50 per cent of the technical poten-tial, or 2000TWh per year.

Utilization of predominantly coastal marine areas that arecurrently regarded as technologically accessible would offera global total potential of approximately 9000TWh per yearfrom wind, waves, currents and osmosis, with wind poweroffering by far the biggest potential and quickest imple-mentation. However, the issue of concurrent utilization ofcoastal marine areas through systems for power generationfrom wind and waves would have to be examined in moredetail, because certain wave energy systems may be difficultto combine with wind farms. In addition, high-density instal-lation of large numbers of systems would lead to significanthabitat changes, e.g. through noise emissions, increasedshipping traffic and other effects such as underwater cables,so that the overall effect caused by concurrent utilization ofseveral technologies have to be regarded as unsustainable.

Renewables vs. CO2 sequestrationGenerating costs for fossil power plants currently rangebetween US$25 and 55 per MWh, for wind energy andsmall-scale hydroelectricity between US$35 and 90 perMWh (IEA, 2005). The additional costs for CO2 sequestra-tion relating to electricity production in fossil power plantsrange between 30 and 60 per cent, depending on the tech-nology and underlying conditions. Assuming moderatefuture fuel price increases and further cost reductions forinvestments in both fossil power plants and renewables,sequestration with continued utilization of fossil energy car-riers would very likely result in higher CO2 avoidance coststhan utilization of renewables in the medium and long term.

In addition, sequestration does not reduce dependenceon fossil fuels and the potential for associated conflicts.Compared with CO2 sequestration, intensive utilization ofrenewables is therefore regarded as the preferred option.


5.3.3.2UNFCCC and Kyoto Protocol

The production of the national emissions inventoriesin accordance with the United Nations FrameworkConvention on Climate Change and the Kyoto Pro-tocol is based on the IPCC Guidelines for NationalGreenhouse Gas Inventories. At present, theseGuidelines do not deal explicitly with the issue ofsequestration. However, the IPCC Special Report onCSS (IPCC, 2005) provides for the option of applyingthe current framework provisions, principles andmethods to sequestration activities. The Norwegianapproach demonstrates how these general provisionscould be applied to sequestration in practice: Norwayreports the quantities of CO2 sequestered at the off-shore Sleipner facility (Box 5.3-1) and consistentlyfactors any emissions leaked during the injectionprocess into its national emissions (IPCC, 2005). Thesequestered CO2 is not added to the emissions inven-tory but is treated, in effect, as non-emitted. TheGuidelines are due to be revised in 2006. It is likelythat the current debate about standards for inventor-izing sequestered CO2 will flow into this process andrelevant provisions will be adopted in the nearfuture. Apart from the practical question of how toinventorize sequestered CO2 in the national reports,a further issue to be clarified is whether, and how,sequestration projects should be integrated into theflexible mechanisms – emissions trading, the CleanDevelopment Mechanism (CDM) and Joint Imple-mentation (JI) (Bode and Jung, 2005; IPCC, 2005).The inclusion of sequestered CO2 in the flexiblemechanisms raises a variety of issues (Bode andJung, 2005) which shall not be discussed in detailhere. Matters become especially complicated in rela-tion to the CDM if, for example, an Annex B country‘imports’ CO2 from developing countries which hasbeen emitted on land and stores it in sub-seabedreservoirs which are already in use. Strictly speaking,such cases do not meet the CDM’s additionality cri-terion, which means that in essence, no CDM emis-sion credits can be issued. Nor does it necessarily pro-mote technology transfer to developing countries,which is an explicit objective of the CDM. Similarlycomplex issues arise in relation to emissions tradingand JI.

5.3.3.3Instruments to regulate CO2 storage in the seabed

WBGU considers that in view of the leakage risk,regulations are required for activities aimed at thestorage of CO2 in the seabed. Firstly, more rigorousminimum standards are needed, with mandatory

compliance in order to minimize risks. Secondly, theuse of quantity restrictions or liability-based instru-ments as a response to the risk of leakage would helpensure that lower-risk sustainable emissions avoid-ance options (e.g. increasing energy efficiency andthe use of renewables) are not neglected.

Geological and technological minimumstandards The rate of CO2 leakage over the long term must bevery low and must also be readily monitored and ver-ified. Firstly, the retention period for stored CO2 atthe chosen site must be very long – at least 10,000years. Our current state of knowledge indicates thatit is possible to meet this criterion, at least in deeperaquifers (Ploetz, 2003; IPCC, 2005). Secondly, theCO2 storage sites must be easily monitored, i.e. itmust be possible to record both the leaked and thesequestered amount of CO2 on a reliable basis. Atpresent, however, adequate technologies to measureCO2 leaks are not available.

Indirect quantity restrictions The leakage risk in particular indicates thatsequestered CO2 cannot be viewed as fully ‘avoided’CO2 emissions in international climate agreements.In the setting and implementation of emissionsreductions targets, storage should therefore only beeligible in part as avoided emissions. Variousapproaches can be considered in this context, both atinternational level (UNFCCC etc.) or solely forEuropean climate policy at first. In the following,WBGU outlines various instruments which aim torestrict by indirect means the proportion of CO2 stor-age. It offers an overview of possible approacheswhich could play an important role in relation to sub-seabed storage as well as sequestration in general. Noconclusive evaluation of the instruments can beundertaken here, firstly because no policy decisionhas been taken on appropriate limitation targets, andsecondly because there is still a considerable need forresearch in many areas (Bode and Jung, 2005; IPCC,2005).• Adding leakage to total emissions: Sequestered

CO2 would only be partly recognized as avoidedemissions. The percentage of CO2 which would beconsidered as having been emitted ‘in practice’and which would have to be designated as such inthe national reports would be determined at polit-ical level. However, this percentage should notmerely reflect but should significantly exceed theprobable leakage, in order to take appropriateaccount of the impacts of leakage on the marineenvironment.

• Deductions in the context of the flexible mecha-nisms: Emission rights arising from sequestration

84

85Recommendations for action: Regulating CO2 storage 5.4

could only be traded with substantial deductions.This would mean that a certificate based on onetonne of sequestered CO2 would give rise to anemissions entitlement of less than one tonne ofCO2. The same principle would apply to CDMcredits from sequestration activities in developingcountries. CDM credits could also be granted forthe storage of ‘imported’ CO2 from developingcountries, especially as this type of cooperativeproject would reduce global CO2 emissions intothe atmosphere – albeit without satisfying the cur-rent CDM criteria (Section 5.3.3.2), which is why ahigher deduction would be justified. The decisionwhich deduction rate would be reasonable in indi-vidual cases would largely depend on the climatepolicy assessment of the leakage risk and theimpacts on marine ecology.There is a considerableneed for further research in this area.

• ‘Traditional Action’: Countries would agree tomeet a specific proportion of their emissionsreduction commitments without recourse to sub-seabed or any other form of CO2 sequestration.This approach would be analogous to the conceptof ‘domestic action’.

Liability mechanisms When applying the above-mentioned instruments tolimit CO2 sequestration, countries implicitly maketheir own assessment of the scale of the leakage riskand the likely damage that leakage would cause. Bycontrast, liability mechanisms are an alternative orsupplementary approach relying on the marketmechanism. An effective liability regime forsequestered CO2 means establishing a transparentand credible system to determine who is responsiblefor discharged CO2 and who is therefore liable to paycompensation: either through ex post adding to over-all emissions, ex post acquisition of emission rights, orpenalty payments which are used for climate andocean protection. As long as the operator still exists,it may be comparatively easy to enforce liability.However, the long time scale of climate protectionmeans that the issue of liability must be clarified andsafeguarded over the long term.The issues surround-ing the cleanup of contaminated sites at nationallevel have shown that it is often the state which ulti-mately shoulders the financial burden. This appliessimilarly to cases involving private operators, espe-cially if a defunct polluting company has no legal suc-cessor or the successor lacks the resources to paydamages.

‘Carbon sequestration bonds’ have emerged as amarket-based solution in this debate (Edenhofer etal., 2005). Here, a firm which intends to sequestrateor store CO2 has to deposit a sum of money with adesignated authority, equivalent to the amount of

sequestered CO2 multiplied by the CO2 certificateprice (Edenhofer, 2003). The company would obtaininterest for the bond, equivalent to the normal mar-ket rate of interest on long-term bonds.The authority– this could be the Climate Central Bank alreadyproposed by WBGU (WBGU, 2003) – devalues thebond according to the fraction of leaked CO2. Thebalance could be used to pay for emissions preven-tion measures, such as the promotion of renewableenergies, or even the purchase and withdrawal ofemissions rights. In the case of leaks from marine dis-posal sites in particular, the funding of marine con-servation measures from these resources would bejustified. As the value of the bond falls, the interestpaid also decreases. No fixed price for the devalua-tion of the bond is set in advance; instead, the deval-uation increases over time in line with the actualamount of leaked CO2. The idea is that the companytries to sell the right to the interest accumulating onthe deposit as a ‘bond’ on the financial markets. Thiscan only be achieved if potential purchasers areoffered a rebate on the value of the bond which ishigh enough to offset the risk of devaluation by theauthority. During trading, the market value wouldreflect not only the devaluation of the depositedamount but also the capital market’s assessment ofthe likely leakage risk in future. The concept of ‘car-bon sequestration bonds’ is a very interesting andinnovative approach to risk assessment and liability,and merits further research.

5.4Recommendations for action: Regulating CO2

storage

5.4.1Prohibiting CO2 injection into the ocean

WBGU firmly rejects the storage of CO2 in theocean, i.e. in the water column and on the sea floor.The ocean is in permanent exchange with the atmos-phere, with the result that this option does not miti-gate the long-term consequences of CO2 emissionsfor future generations. It is therefore not a sustain-able option. The risk that ecosystems will sufferappreciably under an elevated CO2 concentration inthe water is a further argument against the disposalof this greenhouse gas in seawater (Section 5.2.2;IPCC, 2005; Pörtner, 2005). Moreover, the interna-tional community will scarcely be able to control CO2

lakes on the sea floor, and the release of this CO2 tothe atmosphere over the long term cannot beexcluded. WBGU therefore recommends a full and


comprehensive ban on CO2 injection into the ocean,regardless of the territorial status of waters.

The 1972 London Convention on the Preventionof Marine Pollution by Dumping of Wastes andOther Matter, in conjunction with its London Proto-col (Section 5.3.3.1; the protocol has not yet enteredinto force) prohibit in principle the placement of CO2

in the ocean. Both agreements, however, contain animportant exception that needs to be firmly rejectedin light of the above: both permit in their currentwording the injection of CO2 that arises from the pro-duction of mineral oil or natural gas, as long as theassociated processing operations take place at sea.The prohibition on the injection of CO2 arising fromprocessing operations on land that is already implic-itly in place should therefore be extended explicitlyto such CO2 that is separated in the course of seabedresource exploration and processing operations atsea. Such a prohibition could possibly be comple-mented by a corresponding arrangement under theFramework Convention on Climate Change; thiscould also serve to cover those states that do not rat-ify the London Protocol.

5.4.2Limiting CO2 storage in the seabed

The disposal of CO2 in the seabed poses substantiallyless risk than its injection into the water column or onthe sea floor. For that reason, and in view of thealmost unavoidable rise in energy consumption espe-cially in developing and newly industrializing coun-tries,WBGU considers it acceptable for a transitionalperiod to use injection into the geological sub-seabedas an option complementing more sustainable emis-sions reduction strategies.

WBGU accordingly recommends clarifying theissue of conformity of sub-seabed geological storagewith the London Convention or London Protocol inthe relevant bodies of the convention and protocol insuch a way that CO2 sequestration in sub-seabed geo-logical formations is permissible regardless of thelocation of processing operations. If it should notprove possible to generate consensus on construingthese legal provisions to mean that sub-seabed CO2

disposal is permissible, then modifying or supple-menting the London Protocol should be considered.WBGU also argues that such activities should onlybe permitted from the outset for a limited period,such as several decades.

Before the international law of the sea can be con-strued or supplemented in such a way, universal min-imum technological standards would first need to bedefined and complied with. These need to be devel-

oped specifically for marine transport, for CO2 injec-tion and storage, and for the characteristics and mon-itoring of geological disposal sites. As long as theproblems currently associated with the measurementof CO2 releases persist, WBGU advises applyingexceedingly strict requirements upon geological dis-posal sites.WBGU takes the view that in this respect,too, the London Convention or London Protocolprovides an appropriate framework for setting stan-dards, underpinned by more comprehensive rulesgoverning sequestration activities under the Frame-work Convention on Climate Change.

The IPCC Guidelines play an important role inthis context.These guidelines govern the preparationof national emissions inventories.Their review is cur-rently pending. WBGU shares the view of the IPCCSpecial Report (2005) that the present regulatorystructure, including the flexible mechanisms, can inprinciple also be applied to sequestered CO2.WBGUdoes not consider this expedient in all situations, butdoes regard it as purposeful in the case of CO2 dis-posal in verified sub-seabed geological formations.WBGU recommends, however, that when seques-tered CO2 is integrated into inventories and into theflexible Kyoto mechanisms the risk of leakage betaken into account. This can be done through, forinstance, deductions in emissions trading or fromCDM credits, or through liability rules (Section5.3.3.4).


Risks posed by the use of geologicalformations for CO2 storageThere is a need for further research on the perma-nence of marine CO2 storage in deep geological for-mations. The associated monitoring procedures alsoneed further development. Furthermore, researchshould be conducted on the potential impacts of CO2

leakage upon marine ecosystems and organisms.The long-term effects of storage upon atmos-

pheric CO2 concentrations should also be studied.Anissue of particular importance in this respect is whichspecifications a storage site needs to meet in order toensure stable atmospheric CO2 concentrations at alow level over the long term. This will require animproved understanding of the carbon cycle on a mil-lennial time scale.

Legal settingThe instruments of international law governing thepermissibility of CO2 storage in deep sub-seabedgeological formations need to be studied compre-hensively. Not only the London Convention with its

86


1996 Protocol should be taken into account. It isequally important to analyse links to other regimes ininternational law – notably the Framework Conven-tion on Climate Change with its Kyoto Protocol, andthe United Nations Convention on the Law of theSea (UNCLOS).

Regulating CO2 storage in the seabed The manner in which geological storage of CO2 in theseabed (and, it is worth noting, on land) may be eligi-ble as a climate mitigation measure under the inter-national climate protection regime needs to be clari-fied unequivocally in the near future. There is a needfor research in the social sciences and economics onthe issues surrounding the flexible mechanisms. Iden-tifying which instruments for the limitation ofsequestration are effective, efficient and enforceablein international law and policy is an issue of particu-lar importance.

Marine renewables Great uncertainties still attach in some instances tothe renewable energy potential of marine sourcessuch as offshore wind, wave energy, salt gradientenergy or ocean thermal energy conversion. There isa considerable need for further research in order toidentify the sustainable global potential. This con-cerns both the methods and the impacts that need tobe taken into account.

Methane hydrates in the sea floor

Large quantities of carbon are stored in the sea floorin the form of methane hydrates, with an order ofmagnitude comparable to the global occurences ofcoal. There are risks associated with methanehydrates due to climate change as well as ocean min-ing. There are, however, considerable uncertaintiesand gaps in knowledge, so that only a preliminaryevaluation of these risks is possible.

6.1The methane hydrate reservoir

Gas hydrates – such as methane hydrates – are solidscomposed of water molecules that have gas mole-cules enclosed within their crystal lattices. Carbondioxide, hydrogen sulphide and methane moleculeshave the right size to be trapped inside such a hydratecage. Methane hydrates look like dirty ice and areflammable. They store large quantities of methanewithin a very small space: in the transition to the gasphase their volume increases by a factor of 170.

They are only stable under specific pressure andtemperature conditions.The higher the ambient tem-perature, the higher the pressure has to be to preventthe methane hydrate from dissolving. The optimalconditions are typically found on the sea floor atwater depths of at least around 500m, and in the Arc-tic starting already at lower water depth. Here,methane hydrate can form in the sediments providedsufficient quantities of methane are produced by thedecomposition of organic carbon deposits. The car-bon for the methane hydrate is ultimately derivedfrom the biological production of the ocean, as deadbiomass is deposited in the sediments and bacteriallydecomposed on the sea floor (‘biogenic’ methane).The formation of methane hydrates takes a very longtime, so they cannot be considered as a renewableenergy source: the present deposits have probablybeen formed over a period of several million years(Davie and Buffett, 2001). An additional, smallerhydrate source is found in leaking natural gas forma-tions (‘thermogenic’ methane) from which methanebubbles rise through the sediments and under

favourable conditions (i.e., in the hydrate stabilityzone in the cooler upper sediment layers) formhydrates with water.An example can be found in theGulf of Mexico.

As the temperature in the sediment quickly riseswith increasing depth due to the Earth’s heat (ataround 30°C per kilometre) but the pressure – alsoincreasing – cannot compensate for the temperatureincrease, methane hydrates in marine sediments areonly stable down to a certain depth in the sediments.Below the limit of this stability zone, typically severalhundreds of metres thick, methane can again occur asa gas in the sediments.

Gaining evidence for the presence of methanehydrates, directly by drilling or indirectly with seismictechniques, is difficult. While the drilling that hasbeen carried out up to now does not allow broad-area mapping of its occurrences, seismic methods canonly identify the lower limit of the stability zone. Onthis basis, no conclusive statement can be made aboutthe quantity of methane in sediments, because thevolume of the hydrate remains unknown.These mea-surement problems mean that models must be usedto estimate the global reservoir of methane hydrates.In the 1990s it was assumed that carbon quantities onthe order of 10,000Gt C were stored in the form ofmethane hydrates (that equates to around twice theentire fossil energy resource: Rogner, 1997), but cur-rent estimates suggest a much lower value(500–3000Gt C: Buffett and Archer, 2004; Milkov,2004). Klauda and Sandler (2005) presume that thelargest hydrate occurrences are in the deep-seabasins rather than on the continental margins. Theytherefore also report a much higher estimation of78,000Gt C, but this is based on unrealistic assump-tions of the sedimentation rates of organic carbon inthe deep sea. WBGU considers the estimate of500–3000Gt C to be reliable. A comparable amountof methane is present again below the hydrates in thegaseous state (Archer, 2005). Here are some figuresfor comparison: at the end of 2004 the proven coaland natural-gas reserves amounted to 900Gt C and,respectively, 92Gt C (BP, 2005); the atmosphere con-

6

6 Methane hydrates in the sea floor

tains 805Gt C of carbon dioxide, of which 210Gt Cstem from anthropogenic emissions.

6.2Methane release due to human intervention

The stability of methane hydrate deposits can beaffected on the one side by global warming; on theother side, however, there are risks of an uninten-tional release of methane associated with the pro-duction of oil, natural gas, and possibly in the futureof methane hydrate itself.

6.2.1Response to pressure and temperature changes

Changes of pressure and temperature in the hydratelayer lead to changes in the stability zone, i.e., thedepth interval in the sediment where methanehydrate is stable. Higher pressure stabilizes themethane hydrate, while warming reduces the thick-ness of the stability zone. Due to warming, methanehydrate will normally thaw from below (Fig. 6.2-1).Figure 6.2-1a uses a phase diagram to illustrate thestability zone in the ocean and in the underlying sed-iments. The red curve indicates temperature: in theocean it decreases with increasing depth, and in thesediments it increases again due to the Earth’s inter-nal heat. The black curve shows the temperaturebelow which methane hydrate is stable, as deter-mined by the ambient pressure conditions. Thismeans that methane hydrate can only exist in sedi-ments within the depth interval where the actualtemperature (red) is below the stability temperature(black). So the point where the two curves cross in

the sediment represents the lower boundary of thestability zone.

If the ocean warms by 3°C, then the red tempera-ture curve shifts by the corresponding amount to theright (Fig. 6.2-1b). The new point of intersection ofthe temperature and stability-temperature curvesdefines the new lower boundary of the stability zone,which has shifted upward. The amount of gaseousmethane below the hydrate layer has also increasedby the corresponding amount.

Figure 6.2-1c assumes that the ocean rapidlywarmed by 8°C, so that the temperature curve iscompletely to the right of the stability-temperaturecurve, and therefore hydrate is no longer stable atany depth. Whereas with a 3°C ocean-temperatureincrease the total sediment depth down to the base ofthe stability zone first has to warm before themethane hydrate begins to dissolve at all, in theexample with an 8°C increase the destabilization ofthe hydrate would begin at the sea floor, i.e., beforethe total sediment layer has warmed. In the course ofthe temperature rise the methane hydrates woulddissolve completely from above.

6.2.2Effects of climate change on methane hydrates

Global warming leads to temperature changes in theocean as well as to changes in sea level, and thereforeto pressure changes on the sea floor. Figure 6.2-2 pro-vides an overview of the effects this can have onmethane hydrate deposits.

In pessimistic IPCC scenarios the average sea-sur-face temperature increases by the end of this centuryto 5°C above the pre-industrial level. Regionally, forexample in the Arctic, this value could be as great as10°C. The high latitudes are of global importance

90

Methane hydrate

Methane gas

Methane-free sediment

Sea floor

Ocean

a b c

+3°C +8°C

Dep

th

Temperature

FFiigguurree 66..22--11Changes in the methane hydrate layer due to warming. The black curve describes the stability temperature dependent ondepth. The red curve shows the actual temperature; red dashed lines show schematic temperature profiles after a warming of3°C (stability zone of hydrates becomes thinner from the bottom) and 8°C (stability zone completely disappears), respectively.Source: WBGU

91Methane release due to human intervention 6.2

because it is here that the cold-water masses origi-nate that fill the deep sea worldwide. Because of thestable temperature layering and the slow mixing ofthe ocean, the warming, as a rule, will only penetrateto the sea floor very slowly, over the course of severalcenturies. Similar time frames are necessary in orderto warm the sediment layers down to several hun-dreds of metres. Only under very special local condi-tions – with hydrate occurrences at shallow seadepths and in well-mixed marine regions – couldhydrates become unstable in the short term (withinthis century) due to warming. An escape of hydrateson a large scale (that is, enough to have a noticeableimpact on climate) is not an acute but a long-term

danger. Over a period of centuries a reinforcing feed-back loop with global warming could occur, whichover time could become increasingly difficult tocheck.

Relatively rapid and intense local temperaturechanges could occur when marine currents arealtered, a danger that is commonly discussed withrespect to the northern Atlantic (Section 2.1.3). Thedevelopment of temperature at the sea floor seems todepend strongly on how the circulation changes(Mignot et al., submitted) and is therefore difficult topredict. Simulations suggest, however, that after abreakdown of the deep-water formation in the Nor-wegian Sea the bottom temperature in some regions

Certaineffect

Possibleeffect

100

100

100

1,000

1,000

1,000

1,000

10

Amplification of thegreenhouse effect

Global warming

CO2 uptakeby the ocean

Increase of theatmospheric CO2

concentration

Acidification

Submarinelandslides

AtmosphericCH4 oxidizes to CO2

Increase in theatmospheric CH4

concentration

Abrupttemperature change

on the sea floor

Warming of thesediment layer

Increase in thedissolved inorganiccarbon in the ocean

Oxidation of methanerising through the

water column

Warming of the totalocean down to

the sea floor

Methane escapethrough diffusion

Dissolution of hydratesfrom top downward

Methane escapesby blowout

Change inocean currents

Dissolution of hydratesfrom the bottom upward

Warming of thesea surface

FFiigguurree 66..22--22Causes and effects of methane hydrate destabilization. The mechanisms are discussed in the text. Numbers above the arrowsindicate the respective time scale of the process in years (no number given = immediate effect).Source: WBGU

of the North Atlantic could quickly rise by over 7°C.Changes at this order of magnitude could then alsodestabilize hydrate reservoirs.

An additional factor is the rising sea level, which,by increasing the pressure on the sea floor, could inprinciple stabilize the hydrate deposits. Here only thevolume of water released by melting land ice massesis relevant because thermal expansion would notincrease the pressure. The effect, however, is verysmall: in water depths of 400m a pressure increase of0.04MPa (corresponding to a sea-level increase of4m) results in an increase of the destabilization tem-perature of less than 0.1°C. The long-term sea-levelrise can therefore not compensate for the effect ofthe long-term warming on hydrate stability.The sameis true for short-term relative changes in sea levelresulting from circulation changes (Levermann et al.,2005), the results of which cannot compensate for theabrupt temperature changes they also cause.

If the methane hydrate stability zone is reduced,then methane gas forms below the hydrate layer.Thisgas can either penetrate through the hydrate layerand escape out of the sea floor through small chan-nels or permeable sediment layers, or it can blastthrough the hydrate layer if sufficient quantities ofgas collect below a continuously thinning layer. Insuch a blowout large amounts of methane gas areabruptly released. Because the shattered blocks ofmethane hydrate released are less dense than water,they rise to the surface and dissolve there.The quantity of methane gas that would escape fromthe hydrate layers by one of these mechanisms in thefuture can presently only be roughly estimated,because the stability and permeability of sedimentlayers are dependent on highly variable local condi-tions.

6.2.3Mining of methane hydrates

Methane hydrates represent a source of fossil fueland can therefore be of interest for commercialexploitation. The economic feasibility of their recov-ery depends greatly on the available methane con-centration in the hydrate. The few examples of prac-tical experience obtained in exploiting methane fromhydrate deposits are from the Messoyakha gas field(Siberia) and the Mallik (Alaska) research project.The Russian Messoyakha gas field is an occurrencebelow permafrost that was discovered as early as the1960s. Not only were the mining costs here extremelyhigh, but it has also come into question whether themethane recovered here in the 1970s really was, asclaimed, retrieved from hydrate deposits (EIA, 1998;Schindler and Zittel, 2000a). Mallik 2002 is a drilling

project on the Arctic coast of Canada, where themethane concentration of the hydrate is rated similarto that found in Japanese coastal waters. The projectincluded gas hydrate production tests and is part ofan international research consortium in which states(incl. USA, Japan, India and Germany) and compa-nies are participating.

In principle, the mining of methane hydrates onthe high seas would be possible. It is considered tech-nically feasible to drill into the sea floor in waterdepths up to four kilometres.The technical and espe-cially the economic practicability of potential recov-ery mining methods is a subject of research in whichJapan and the USA are playing particularly impor-tant roles. The Japanese programme for methanehydrate mining (National Methane HydrateExploitation Program, MH21), among other aspectsof methane hydrate research, is expressly pursuingthe ambitious goal of beginning production tests in2007 and is aiming to have the technology for com-mercial large-scale production by 2012 (MH21,2005). Financing for the US American methanehydrate research programme (Methane HydrateResearch and Development Act of 2000) wasextended through 2010 by the Energy Policy Act of2005. Commercial mining of methane hydrate in USAmerican waters is deemed possible by 2015 andlarge-scale mining by 2020 (DOE-NETL, 2005; Ray,2005).

These expectations are compatible with the esti-mation that methane hydrate mining will be eco-nomically feasible in some regions within the next5–10 years, while it would take 30–50 years beforeworldwide massive mining is possible (MethaneHydrate Advisory Committee, 2002; Collett, 2005).Methane hydrate exploitation in permafrost areas onland could reach industrial proportions more quicklythan the exploitation from the sea (Johnson, 2004).That is because progress in the identification andevaluation of occurrences feasible for exploitationon land is ahead of that for occurrences beneath thesea. In addition, there has already been extensiveexperience gained in recovery and production tech-nology on land (Mallik research drilling, Messoyakhagas field). The more favourable recovery conditionscompared to the sea also make it likely that miningwill first be carried out on land. In combination witheconomies of scale and learning effects, there couldtherefore be cost advantages. Overall, this means thatthere is an initial advantage for methane hydrateexploitation on land over that at sea. The predictedtechnological feasibility as well as the economic andenergy-strategic potential of this kind of energy pro-duction, however, is critically questioned and consid-ered to be clearly overestimated (Schindler and Zit-tel, 2000b).

6 Methane hydrates in the sea floor92

93Possible results of methane release 6.3

Targeted research into the production of marinemethane hydrate has been limited so far to a fewpilot studies. They probably will not go beyond thestage of feasibility studies during this decade.

6.3Possible results of methane release

The consequences of a release of methane gas fromhydrates depend on the mechanism – ‘diffusion’ or‘blowout’ – as well as the time scale of the release.

When methane gas diffuses through the hydratelayer and slowly escapes in small bubbles from thesea floor, a large portion of it will probably be dis-solved in the water column as it rises. A new studyshows, however, that methane bubbles could alsopossibly rise through the upper water layers andescape into the atmosphere (Sauter et al., 2006). Dis-solved methane in the ocean has a lifetime of about50 years before it oxidizes to H2O and CO2. A largeportion of the released methane would therefore bereleased to the atmosphere before it oxidizes. Firstly,the remaining oxidized portion would increase theconcentration of dissolved inorganic carbon in theocean, which contributes to further acidification(Section 4.1). Secondly, an equivalent decrease inoxygen concentration would occur. For comparison:in order to exhaust all of the 2 · 1017mol of oxygencontained in the ocean, it would have to react with1000Gt of methane (Archer, 2005). Thirdly, in thelong term, a new carbon-equilibrium state would beestablished between the atmosphere and ocean, overthe course of about 1000 years, and about one-fifth ofthe carbon incorporated in the ocean released into

the atmosphere. The concentration of CO2 in theatmosphere would thereby increase, strengtheningthe greenhouse effect. Hence, over the long term, thiseffect would come about in any case: the result is thesame whether methane escapes directly into theatmosphere and oxidizes there to CO2, four-fifths ofwhich is gradually taken up by the ocean, or if it isfirst released in the ocean, oxidized there, and one-fifth is given off to the atmosphere.

When large quantities of methane are suddenlyreleased, most of it will reach the water surface andabruptly increase the methane concentration in theatmosphere. Because methane is a considerablymore effective greenhouse gas than CO2 (around 25times stronger per molecule) due to its much lowerconcentration and therefore less saturated absorp-tion bands, the effect of comparatively low amountsof methane is significant. But atmospheric methanequickly oxidizes (with an average residence time ofeight years), to CO2, which accumulates in the atmos-phere due to its long life expectancy, so that in thelong term the escaped methane after its oxidation toCO2 has an even greater impact on climate thanbefore.

Figure 6.3-1 shows how anthropogenic CO2 emis-sions can lead to methane emissions from hydratedeposits over the coming millennia. A total emissionof 1000Gt CO2 is assumed. Figure 6.3-1a reveals howstrongly this could cause the atmospheric methaneconcentration to increase, whereby the uncertainty ofthe time scale of the release is taken into accountwith three different assumptions.

Figure 6.3-1b illustrates the climatic consequencesof the methane emissions for the 1000Gt of CO2 sce-nario for the case of a methane release within 1000

4

3

2

1

0

Atm

osp

heric

CH

4 [p

pm

]

200 40 60 80 100Millenia

a

1,00010,000100,000

Duration of release [Years]4

3

2

1

0200 40 60 80 100

Millenia

b

Rad

iativ

e fo

rcin

g [W

/m2 ]

Total CO2 increase

CO2 increase due toanthropogenic emissionsCH4

FFiigguurree 66..33--11(a) Atmospheric methane concentration for a scenario with a total quantity of 1000Gt of anthropogenic CO2 emissions. Thecurves describe the resulting methane release over different time frames (1, 10, and 100 thousand years). (b) Climate-impactingradiative forcing for the case of the shortest release period of 1000 years. This is a combination of the forcing due to methaneitself (green; it gradually oxidizes to CO2 and thus disappears), that due to anthropospheric CO2 emissions (black), and CO2

from the oxidation of methane. The last two together yield the radiative forcing due to the total increase of CO2 (red).Source: Archer und Buffet, 2005

6 Methane hydrates in the sea floor

years. The results are caused both directly throughthe increase in atmospheric methane concentration(green), as well as on a longer time scale by theincrease of the CO2 concentration (red). Althoughthe direct methane effect is lower than that of theoriginal anthropogenic CO2 emission, the subsequentincrease in CO2 concentration through oxidation ofthe methane leads, over the long term, to a near dou-bling of the greenhouse effect.

Methane eruptions can also present other dan-gers.They can destabilize continental slopes and trig-ger large submarine landslides, which can then possi-bly result in the break-up of additional hydrates. Evi-dence of such slides can be found on the sea floor. Forexample, in the Storegga landslide off the coast ofNorway around 8000 years ago, an average of 250mof the continental slope with a width of 100km weretransported downslope (Archer, 2005). This eventtriggered a tsunami that was at least 25m high off theShetland Islands and at least 5m high along theBritish coast (Smith et al., 2004). The amount ofmethane released by this landslide is estimated at0.8Gt C (Archer, 2005). When this amount ofmethane directly enters the atmosphere, it can alterthe radiative forcing by as much as 0.2W per m2 (forcomparison, today’s radiative forcing due to anthro-pogenic greenhouse gases is 2.7W per m2). Thisexample illustrates that an abrupt release ofmethane, even in the case of a large catastrophic slideof the continental slope, would only have a relativelyminor impact on climate.

6.4Recommendations for action: Preventing methanerelease

Through the warming of seawater, anthropogenic cli-mate change can lead to a destabilization of methanehydrate deposits on the sea floor. According to thepresent state of knowledge, however, the danger of asudden release of large, climate-influencing quanti-ties within this century is very small. Of much greaterimportance is the probability of a continuousmethane release over many centuries to millenniadue to the slow intrusion of global warming into thedeeper ocean layers and sediments. The conse-quences of human actions persist in this respect notjust over centuries, but could influence the Earth’sclimate over tens of thousands of years.

Limiting global warming follows once more hereas a recommendation for action, because methanerelease from hydrates could further amplify climatechange in the long term. This feedback effect pre-sents the danger that humankind could lose controlof the greenhouse-gas concentration in the atmos-

phere, as the outgassing of methane from the seafloor cannot be controlled or limited.

There is already a need today for institutionalaction with regard to marine methane hydratedeposits. This is with respect to, for one, the targetedmining of marine methane hydrates, and for anotherto the unintentional release of methane that couldoccur during sea-floor mining.

Theoretically, efforts to recover methane fromhydrates could unintentionally trigger their releaseinto the environment, in the worst case as a suddeneruption. The risks of this have not yet been suffi-ciently investigated (Archer, 2005). A leak ofmethane into the environment during mining wouldunnecessarily amplify global warming. In the worstcase even a slope slide could be caused that couldtrigger a tsunami.

The risks associated with mining are very variabledepending on the geological conditions. The risks ofmethane mining therefore have to be carefullyreviewed for each individual case. An environmentalimpact assessment along with monitoring accordingto universal standards is necessary for every case.The International Seabed Authority, an institution ofthe international United Nations Convention on theLaw of the Sea (UNCLOS), is responsible formethane hydrate deposits as well as for otherresources on the sea floor outside the exclusive eco-nomic zone.The Authority grants mining licenses andmonitors mining operations. Its regulations adoptedin 2000 for the exploration of deep-sea mineralresources contain various environmental aspects.This is a starting point for agreement on concretestandards for mining marine methane hydrate on thehigh seas. In the opinion of WBGU it is furthermorenecessary to improve and expand the monitoring sys-tem. It is, however, important to note here that so far‘only’ about 150 countries have ratified UNCLOS,and of those only about 120 countries have ratifiedthe rules governing seabed resources (those whohave not signed include, for example, Iran and theUSA). A framework within which more countriescan be persuaded to accede to the agreements formaintaining universal standards in hydrate miningstill needs to be worked out. Also needed are agree-ments binding under international law for the miningof methane hydrates in marine regions that lie withinthe territorial sovereign rights of coastal nations(Box 2.6-1). This is necessary considering that boththe above-mentioned Japanese pilot project andAmerican plans target future commercial methaneproduction from hydrate deposits in national coastalwaters.

The danger of methane hydrate release also existsin principle in other sea-floor mining activities. Ifmethane were to be destabilized and unintentionally

94


released in the mining of resources, these emissionswould hardly be measurable, and therefore notaccounted for in the emissions inventory of a country,or only insufficiently so. The applicable IPCC guide-lines of 1996 for national emissions inventories donot include methane that is unintentionally emittedat sea.WBGU therefore recommends for the upcom-ing reworking of the guidelines in 2006 that this omis-sion be corrected despite the difficulties in measure-ment. But at the very least a reporting obligationshould be introduced for such releases of methane.


Because estimates of the risks of methane release arestill hampered by large uncertainties and gaps inknowledge, there is a significant need for research.Tobegin with, the methane occurrences need to be moreextensively mapped and quantities estimated. Theprimary focus here should not be on the potentialworkable deposits, but on occurrences that couldpossibly become destabilized by climate change, andon the danger of slope slides. Furthermore, modellingstudies should be employed to investigate whichregions of the ocean show the greatest risk forhydrates to become destabilized through globalwarming.

While research on the long-term stability ofmarine methane hydrates and climate protectionimplications should continue to be strengthened,WBGU sees no need for government subsidies forapplied research for the mining of marine methanehydrates. Public funding of such projects does notseem purposeful because mining poses considerablerisks and methane hydrates do not represent a sus-tainable energy source.

There would, however, be a need for targeted nat-ural science research if appropriate standards for themining of marine methane hydrates need to bedefined. Natural science investigations should besupplemented by social science and legal studies ofthe prospects for worldwide implementation of suchstandards.

Core messages

Climate mitigation for marine conservation

The future of the marine environment will dependcrucially upon whether human-induced disruption ofthe climate system can be limited to a tolerable level.It follows in WBGU’s view that global anthropogenicgreenhouse gas emissions will need to be approxi-mately halved by 2050 from 1990. Because of the geo-physical time lags, the climate protection policiesadopted in the next few decades will determine thestate of the oceans for millennia to come.Adaptationmeasures can only succeed if the present accelerationof sea-level rise and the increasing acidification ofthe oceans are halted.

WBGU has already formulated a ‘climate guardrail’ in earlier reports as a contribution to making asustainable development pathway operable: to pre-vent dangerous climatic changes, the mean global risein near-surface air temperature must be limited to amaximum of 2°C relative to the pre-industrial valueand the rate of temperature change must be limitedto a maximum of 0.2°C per decade. The presentreport shows that marine conservation is a furtherreason why it is essential to obey this guard rail.

Bolstering the resilience of marine ecosystems

Fish stocks and coral reefs will only retain their pro-ductivity and diversity if sustainable marine resourcemanagement is ensured worldwide. The mountingdirect and indirect pressures generated by anthro-pogenic greenhouse gas emissions are making adop-tion of an ‘ecosystem approach’ for the conservationand use of the marine environment ever more impor-tant. To that end, the establishment of marine pro-tected areas, an approach already agreed by theinternational community, must be pushed forwardenergetically, and the regulatory gap for the high seasmust be closed by adopting a corresponding agree-ment within the framework of the United NationsConvention on the Law of the Sea (UNCLOS).

To conserve marine ecosystems and strengthentheir resilience,WBGU proposes the following guardrail: at least 20–30 per cent of the area of marineecosystems should be designated for inclusion in anecologically representative and effectively managedsystem of protected areas.

Limiting sea-level rise and reorienting coastalzone management strategies

The strategies hitherto adopted to protect and utilizecoastal areas are no longer adequate to cope with cli-mate-driven sea-level rise and the mounting destruc-tive force of hurricanes. Novel combinations of mea-sures (portfolio strategies) are called for, wherebythe options of protection, managed retreat andaccommodation need to be weighed against eachother. In particular, coastal protection and natureconservation concerns must be better linked, and thepeople affected by adaptation or resettlement mea-sures must be involved in the planning and imple-mentation of such measures.To this end,WBGU rec-ommends creating integrative institutions that com-bine all key competencies.

To prevent severe damage and losses from occur-ring, and to avoid overstretching the adaptive capac-ity of coastal ecosystems and infrastructure, WBGUproposes the following guard rail for sea-level rise:absolute sea-level rise should not exceed 1 m in thelong term, and the rate of rise should remain below5 cm per decade at all times.

Adopting innovative instruments of internationallaw for refugees from sea-level rise

At present no nation has any obligation under inter-national law to receive migrants whose homeland hasbeen lost due to climate-induced flooding. In the longterm, however, the international community will notbe able to ignore the issue of ‘sea-level refugees’ andwill therefore need to develop appropriate instru-ments for the secure reception of affected people in

7

7 Core messages

suitable areas, ideally in areas that correspond totheir preferences. It would be expedient to develop afair burden-sharing system, under which states makea binding commitment to assume responsibility forthese people in line with their greenhouse gas emis-sions. To inform the policymaking process, studies inthe fields of law and social sciences should be under-taken.

Halting ocean acidification in time

The oceans have absorbed about one-third of allanthropogenic CO2 emissions to date, which hasalready caused a significant acidification (decrease inpH) of seawater. These emissions thus influence themarine environment directly – in addition to theroute via climate change. Unabated continuation ofthis trend will lead to a level of ocean acidificationthat is without precedent in the past several millionyears and will be irreversible for millennia. Theeffects upon marine ecosystems cannot be forecastexactly but profound changes to the food web areconceivable, as calcification of marine organisms maybe impeded or in some cases even prevented.WBGUrecommends fostering internationally coordinatedresearch and monitoring programmes on this issue.Furthermore, the negotiations on future commit-ments under the United Nations Framework Con-vention on Climate Change need to take into accountthe special role of CO2 compared to other green-house gases. Besides stabilizing the overall packageof greenhouse gases, it will be important to also seekexplicitly to stabilize CO2 concentrations.

To protect the oceans against acidification,WBGU proposes the following guard rail: in order toprevent disruption to calcification of marine organ-isms and the resultant risk of fundamentally alteringmarine food webs, the pH of ocean surface watersshould not drop more than 0.2 units below the pre-industrial level in any larger ocean region (i.e. also inthe global mean).

98

References

ACIA (Arctic Climate Impact Assessment) (2005) Arctic Cli-mate Impact Assessment. Cambridge University Press, Cam-bridge, New York.

Adger, W N, Hughes, T P, Folke, C, Carpenter, S R and Rock-ström, J (2005) ‘Social-ecological resilience to coastal disas-ters’. Science 309: pp1036–9.

Agardy, T, Bridgewater, P, Crosby, M P, Day, J, Dayton, P K,Kenchington, R, Laffoley, D, McConney, P, Murray, P A,Parks, J E and Peau, L (2003) ‘Dangerous targets? Unre-solved issues and ideological clashes around marine pro-tected areas’. Aquatic Conservation Marine and FreshwaterEcosystems 13(4): pp353–67.

Ahmed, S, Mbasis, J, Moko, D and Ngonzi, A (2005) ‘Health iswealth: how low-income people finance health care’. Journalof International Development 17: pp383–96.

Alheit, J, Möllmann, C, Dutz, J, Kornilovs, G, Loewe, P,Mohrholz, V and Wasmund, N (2005) ‘Synchronous ecologi-cal regime shifts in the central Baltic and the North Sea in thelate 1980s’. ICES Journal of Marine Science 62: pp1205–15.

Archer, D (2005) ‘Destabilization of Methane Hydrates: ARisk Analysis’. Expertise for WBGU Special Report ‘TheFuture Oceans – Warming Up, Rising High, Turning Sour’.WBGU website, http://www.wbgu.de/wbgu_sn2005_ex01.pdf.

Archer, D (2006) Global Warming – Understanding the Fore-cast. Blackwell Publishers, Oxford, UK.

Archer, D and Buffett, B (2005) ‘Time-dependent response ofthe global ocean clathrate reservoir to climatic and anthro-pogenic forcing’. Geochemistry Geophysics Geosystems 6(3):pp13.

Athreye, V and Roth, J (2005) CGAP Working Group onMicroinsurance. Good and Bad Practices. Case Study No. 14.New Delhi: Tata-AIG Life Insurance Company Ltd.

Atkinson, A, Siegel, V, Pakhomov, E A and Rothery, P (2004)‘Long-term decline in krill stock and increase in salps withinthe Southern Ocean’. Nature 432: pp100–3.

Attrill, M J and Power, M (2002) ‘Climatic influence on amarine fish assemblage’. Nature 417: pp275–8.

Baker,A C, Craig, J, Starger,T, McClanahan, R and Glynn, P W(2004) ‘Corals’ adaptive response to climate change’. Nature430: pp741.

Balmford, A, Gravestock, P, Hockley, N, McCLean, C J andRoberts, C M (2004) ‘The worldwide costs of marine pro-tected areas’. PNAS 10(1073): pp9694–7.

Barber, R (2001) ‘Upwelling ecosystems’. In Steele, J H,Thorpe, S A and Turekian, K K (eds) Encyclopedia of OceanSciences. Academic Press, London, pp3128–35.

Barlow, P M (2003) Ground Water in Freshwater-SaltwaterEnvironments of the Atlantic Cost. Circular 1262. US Geolog-ical Survey, Reston, Virginia.

Barnett, J and Adger, W N (2003) ‘Climate dangers and atollcountries’. Climatic Change 61: pp321–37.

Barrett, P (2003) ‘Palaeoclimatology – cooling a continent’.Nature 421: pp221–3.

Barry, J P, Buck, K R, Lovera, C F, Kuhnz, L, Whaling, P J,Peltzer, E T,Walz, P and Brewer, P G (2004) ‘Effects of directocean CO2 injection on deep-sea meiofauna’. Journal ofOceanography 60: pp759–66.

Baumgartner, T R, Soutar, A and Ferreira-Bartrina, V (1992)‘Reconstruction of the history of pacific sardine and northernanchovy populations over the past two millennia from sedi-ments of the Santa Barbara Basin, California’. CaliforniaCooperative Oceanic Fisheries Investigations (ColCofi) 33:pp24–40.

BBK (Bundesamt für Bevölkerungsschutz und Katastrophen-hilfe) (2006) ‘Schutz kritischer Infrastrukturen’. Bundesamtfür Bevölkerungsschutz und Katastrophenhilfe website,http://www.bbk.bund.de (viewed 15. February 2006).

Beare, D J, Burns, F, Greig,A, Jones, E G, Peach, K, Kienzle, M,McKenzie, E and Reid, D G (2004) ‘Long-term increases inprevalence of North Sea fishes having southern biogeo-graphic affinities’. Marine Ecology Progress Series 284:pp269–78.

Beaugrand, G, Reid, P C, Ibanez, F, Lindley, J A and Edwards,M (2002) ‘Reorganization of North Atlantic marine copepodbiodiversity and climate’. Science 296: pp1692–5.

Beaugrand, G and Reid, P C (2003) ‘Long-term changes in phy-toplankton, zooplankton and salmon related to climate’.Global Change Biology 9: pp801–17.

Beaugrand, G, Brander, K M, Lindley, J A, Souissi, S and Reid,P C (2003) ‘Plankton effect on cod recruitment in the NorthSea’. Nature 426: pp661–4.

Beaugrand, G (2004) ‘The North Sea regime shift: evidence,causes, mechanisms and consequences’. Progress in Oceano-graphy 60: pp245–62.

Bellona Foundation (2005) CO2 for EOR on the NorwegianShelf – the Environment and Economic Growth Hand inHand. Bellona Foundation, Oslo.

Bellwood, D R, Hughes, T P, Folke, C and Nystrom, M (2004)‘Confronting the coral reef crisis’. Nature 429: pp827–33.

Bertrand, A, Segura, M, Gutiérrez, M and Vásquez, L (2004)‘From small-scale habitat loopholes to decadal cycles: a habi-tat-based hypothesis explaining fluctuation in pelagic fishpopulations off Peru’. Fish and Fisheries: pp296–316.

Beyerlin, U (2000) Umweltvölkerrecht. Beck, Munich.

Bijlsma, L, Ehler, C N, Klein, R J T, Kulshrestha, S M, McLean,R F, Mimura, N, Nicholls, R J, Nurse, L A, Perez Nieto, H,Stakhiv, E Z, Turner, R K and Warrick, R A (1996): ‘Coastalzones and small islands’. In Watson, R T, Zinyowera, M C andMoss, R H (eds) Impacts, Adaptations and Mitigation of Cli-mate Change: Scientific-Technical Analyses. Cambridge Uni-versity Press, Cambridge, New York: pp289–324.

8

8 References

Blindheim, J, Toresen, R and Loeng, H (2001) ‘Fremtidige kli-matiske endringer og betydningen for fiskeressursene.Havets Miljø 2001’. Fisken og Havet 2: pp73–8.

Bloomfield, J, Smith, M and Thompson, N (1999) Hot Nights inthe City. Environmental Fund, Washington, DC.

Bode, S, Isensee, J, Krause, K and Michaelowa, A (2002) ‘Cli-mate policy: analysis of ecological, technical and economicimplications for international maritime transport’. Interna-tional Journal of Maritime Economics 4: pp164ff.

Bode, S and Jung, M (2005) Carbon Dioxide Capture and Stor-age (CCS) – Liability for Non-Permanence Under theUNFCCC. Hamburgisches Welt-Wirtschafts-Archiv(HWWA), Hamburg.

Boersma, D, Ogden, J, Branch, G, Bustamante, R, Campagna, C,Harris, G and Pikitch, E K (2004): ‘Lines on the water: ocean-use planning in large marine ecosystems’. In Glover, L K andEarle, S A (eds) Defying Ocean's End.An Agenda for Action.Island Press, Washington, DC: pp123–38.

Bohnsack, J A (1998) ‘Application of marine reserves to reeffisheries management’. Australian Journal of Ecology 23:pp298–304.

Bohnsack, J A, Causey, B, Crosby, M P, Griffis, R B, Hixon, M A,Hourigan, T F, Koltes, K H, Maragos, J E, Simons, A andTilmant, J T (2002) A Rationale for Minimum 20–30% No-Take Protection. Paper Presented to the 9th InternationalCoral Reef Symposium. Bali, Indonesia 23–27 October 2000.Conference Secretariat, Bali.

Bopp, L, Monfray, P, Aumont, O, Dufresne, J-L, Le Treut, H,Madec, G, Terray, L and Orr, J C (2001) ‘Potential impact ofclimate change on marine export production’. Global Bio-geochemical Cycles 15(1): pp81–99.

Boyd, E, Clarke, J, Gray, G, Hodge, G B, Hurlston, L-A, Nichol-son-Cole, S A, Tompkins, E L, Trotz, N and Varlack, L (2005)Surviving Climate Change in Small Islands: A Guidebook.Tyndall Centre for Climate Change Research, Norwich.

BP (British Petroleum) (2005) Putting Energy in the Spotlight.BP Statistical Review of World Energy. BP, London.

Brander, K M (2005) ‘Assessment of Possible Impacts of Cli-mate Change on Fisheries’. Expertise for WBGU SpecialReport ‘The Future Oceans – Warming Up, Rising High,Turning Sour’. WBGU website, http://www.wbgu.de/ wbgu_sg2005_ex02.pdf.

Broecker, W (1987) ‘Unpleasant surprises in the greenhouse?’Nature 328: pp123.

Brooks, N, Nicholls, R and Hall, J M (2006) ‘Sea Level Rise:Coastal Impacts and Responses’. Expertise for WBGU Spe-cial Report ‘The Future Oceans – Warming Up, Rising High,Turning Sour’. WBGU website, http://www.wbgu.de/wbgu_sn2005_ex03.pdf.

Brown, W and Churchill, C F (1999) Providing Insurance toLow-Income Households. Part 1:A Primer on Insurance Prin-ciples and Products. Microenterprises Best Practices,Bethseda, MD.

Brown, W and Churchill, C F (2000) Providing Insurance toLow-Income Households. Part 2: Initial Lessons from Micro-Insurance Experiments for the Poor. Microenterprises BestPractices, Bethseda, MD.

Bruneau, M, Chang, S, Eguchi, R, Lee, G, O’Rourke, T, Rein-horn, A, Shinozuka, M, Tierney, K, Wallace, W and von Win-terfelt, D (2003) ‘A framework to quantitatively assess andenhance the seismic resilience of communities’. EERI SpectraJournal 19: pp733–52.

Bruun, P (1962) ‘Sea level rise as a cause of shore erosion’.Journal of Waterways and Harbours Division ASCE (88):pp117–30.

Bryden, H L, Longworth, H R and Cunningham, S A (2005)‘Slowing of the Atlantic meridional overturning circulation at25ºN’. Nature 438: pp655–65.

Buddemeier, R W, Kleypas, J A and Aronson, R B (2004) CoralReefs and Global Climate Change: Potential Contributions ofClimate Change to Stresses on Coral Reef Ecosystems. PewCenter on Global Climate Change, Arlington.

Buffett, B and Archer, D (2004) ‘Global inventory of methaneclathrate: sensitivity to changes in the deep ocean’. Earth andPlanetary Science Letters 227(3-4): pp185–99.

Bundesregierung (2006) Integriertes Küstenzonenmanagementin Deutschland. Nationale Strategie für ein integriertes Küsten-zonenmanagement. Bundesregierung, Berlin.

Burgess, K and Townend, I (2004) The impact of climate changeupon coastal defence structures. Proceedings of the 39th DefraFlood and Coastal Management Conference. Department forEnvironment, Food and Rural Affairs (Defra), London.

Burns, W C (1997) ‘Global warming - the United NationsFramework Convention on Climate Change and the future ofsmall island states’. Dickinson Journal of Environmental Lawand Policy 6: pp147ff.

Burton, I and Dore, M H I (2000) ‘The Costs of Adaptation toClimate Change in Canada: A Stratified Estimate by Sectorsand Regions. First Deliverable:A Review of the Literature onthe Costs of Adaptation to Climate Change’. NaturalResources Canada website, http://adaptation.rncan.gc.ca/app/filerepository/E74BFBF66A704A778C4D1AD4A8C4BFC2.pdf (viewed 19. January, 2006).

Burton, I and Lim, B (2001) Adaptation Policy Framework.Draft Report. United Nations Development Programme(UNDP), New York.

Burton, I, Huq, S, Lim, B, Pilifosova, O and Schipper, E L (2002)‘From impacts assessment to adaptation priorities; the shap-ing of adaptation policy’. Climate Policy 2: pp145–59.

Caldeira, K, Jain, A K and Hoffert, M I (2003) ‘Climate sensi-tivity uncertainty and the need for energy without CO2 emis-sion’. Science 299: pp2052–4.

Caldeira, K, Akai, M, Brewer, P, Chen, B, Haugan, P, Iwama, T,Johnston, P, Kheshgi, H S, Q, L, Ohsumi, T, Pörtner, H O,Sabine, C, Shirayama, Y. and Thomson, J (2005): ‘Chapter 6:Ocean Storage’. In Intergovernmental Panel on ClimateChange (IPCC) (ed) IPCC Special Report on Carbon Diox-ide Capture and Storage. Cambridge Unviversity Press, Cam-bridge, New York: pp4–41.

Caldeira, K and Wickett, M E (2005) ‘Ocean model predicitionsof chemistry changes from carbon dioxide emissions to theatmosphere and ocean’. Journal of Geophysical Research –Oceans 110: ppC09S4, doi: 10.1029/2004JC002671.

Carlton, J (2005) ‘Is Global Warming Killing Polar Bears?’ TheWall Street Journal via AP website, http://reference.aol.com/article?id=20051216151609990001 (viewed 20.March 2006).

Cazenave, A and Nerem, R (2004) ‘Present-day sea levelchange: Observations and causes’. Reviews of Geophysics42(3): pp139–50.

CBD (Convention on Biological Diversity) (2003) Integrationof Outcome-Oriented Targets Into the Programmes of Work ofthe Convention, Taking into Account the 2010 BiodiversityTarget, the Global Strategy for Plant Conservation, and Rele-vant Targets set by the World Summit on Sustainable Develop-ment. UN Document UNEP/CBD/COP/7/20/Add.5. CBDSecretariat, Montreal.

CBD (Convention on Biological Diversity) (2004a) ProtectedAreas (Articles 8 (a) to (e)). Decision VII/28. UN DocumentUNEP/CBD/COP/7/21. CBD Secretariat, Montreal.

100

101References 8

CBD (Convention on Biological Diversity) (2004b) Marineand Coastal Biological Diversity. Decision VII/5. UN Docu-ment UNEP/CBD/COP/7/21. CBD Secretariat, Montreal.

CBD (Convention on Biological Diversity) (2005a) Ad HocOpen-ended Working Group on Protected Areas:The Interna-tional Legal Regime of the High Seas and Seabed Beyond theLimits of National Jurisdiction and Options for Cooperationfor the Establishment of Marine Protected Areas in MarineAreas Beyond the Limits of National Jurisdiction. UN Docu-ment UNEP/CBD/WG-PA/INF/2. CBD Secretariat, Mon-treal.

CBD (Convention on Biological Diversity) (2005b) Ad HocOpen-ended Working Group on Protected Areas: Options forCooperation for the Establishment of Marine Protected Areasin Marine Areas Beyond the Limits of National Jurisdiction.UN Document UNEP/CBD/WG-PA/1/2. CBD Secretariat,Montreal.

Chape, S, Harrison, J, Spalding, M and Lysenko, I (2005) ‘Mea-suring the extent and effectiveness of protected areas as anindicator for meeting global biodiversity targets’. Philosoph-ical Transactions of the Royal Society 360: pp443–55.

Church, J A, White, N J and Arblaster, J M (2005) ‘Significantdecadal-scale impact of volcanic eruptions on sea level andocean heat content’. Nature 438: pp74–7.

Church, J A and White, N J (2006) ‘A 20th century accelerationin global sea-level rise’. Geophysical Research Letters 33(1):ppdoi:10.1029/2005GL024826.

Chylek, P and Lohmann, U (2005) ‘Ratio of the Greenland toglobal temperature change: Comparison of observations andclimate modeling results’. Geophysical Research Letters 32:L14705.

CIA (Central Intelligence Agency) (2005) ‘The World Fact-book 2005’. CIA website, http://www.cia.gov/cia/publica-tions/factbook/ (viewed 8. March 2006).

Clark, P U, McCabe, A M, Mix, A C and Weaver, A J (2004)‘Rapid rise of sea level 19,000 years ago and its global impli-cations’. Science 304: pp1141–4.

Clarke, T (2001) ‘Taming Africa’s killer lake’. Nature 406:pp554–5.

Claussen, M, Ganopolski, A, Brovkin, V, Gerstengarbe, F-Wand Werner, P C (2003) ‘Simulated global-scale response ofthe climate system to Dansgaard/Oeschger and Heinrichevents’. Climate Dynamics 21(5-6): pp361–70.

Cohen, J E, Small, C, Mellinger, A, Gallup, J and Sachs, J D(1997) ‘Estimates of coastal population’. Science 278:pp1211–2.

Cohen, M C L and Lara, R J (2003) ‘Temporal changes of man-grove vegetation boundaries in Amazonia: application of GISand remote sensing techniques’. Wetlands Ecology and Man-agement 11: pp223–31.

Cohen, M, McCord, M J and Sebstad, J (2005) ‘Reducing vul-nerability: demand for and supply of microinsurance in EastAfrica’. Journal of International Development 17: pp319–25.

Collett,T S (2005) ‘Natural Gas Hydrates: Resource of the 21stCentury?’ American Association of Petroleum Geologists(AAPG) website, http://emd.aapg.org/technical_areas/gas_hydrates/resource_toc.cfm (viewed 22. November 2005).

Commission of the European Communities (2005a) Grünbuchüber ein Programm für den Schutz kritischer Infrastrukturen.KOM(2005) 576 engültig. European Community (EC), Brus-sels.

Commission of the European Communities (2005b) Thematis-che Strategie für den Schutz und die Erhaltung der Meer-sumwelt. Mitteilung der Kommission an den Rat und dasEuropäische Parlament. KOM(2005)504 endgültig.SEC(2005)1290. European Community (EC), Brussels.

CSD (Commission on Sustainable Development) (2004)Review of Progress in the Implementation of the Programmeof Action for the Sustainable Development of Small IslandDeveloping States. Report of the Secretary-General. UN Doc-ument E/CN.17/2004/8. CSD, New York.

CSLF (Carbon Sequestration Leadership Forum) (2005) CO2Capture and Storage Summary of Norway – Paper presentedby Odd Sverre Haraldsen (Director General, Ministry ofPetroleum an Energy) at the 1th International Workshop onCSLF Projects in Berlin, September 27-29, 2005. CSLF, Wash-ington, DC.

Curry, R and Mauritzen, C (2005) ‘Dilution of the northernNorth Atlantic Ocean in recent decades’. Science 308:pp1772–4.

Czisch, G (2005) Szenarien zur zukünftigen Stromversorgung.Kostenoptimierte Variationen zur Versorgung Europas undseiner Nachbarn mit Strom aus erneuerbaren Energien. Doc-toral Thesis. University Kassel, Kassel.

Dahdouh-Guebas, F, Jayatissa, L P, Di Nitto, D, Bosire, J O, LoSeen, D and Koedam, N (2005) ‘How effective were man-groves as a defence against the recent tsunami?’ Current Biol-ogy 15: pp1337–8.

Danielsen, F, Sørensen, M K, Olwig, M F, Selvam, V, Parish, F,Burgess, N D, Hiraishi, T, Karunagaran, V M, Rasmussen, MS, Hansen, L B, Quarto, A and Suryadiputra, N (2005) ‘TheAsian Tsunami: a protective role for coastal vegetation’. Sci-ence 310: p643.

Darwin, R F and Tol, R S J (2001) ‘Estimates of the economiceffects of sea level rise’. Environmental and Resource Eco-nomics 19: pp113–9.

Daschkeit, A and Schottes, P (eds) (2002) Klimafolgen fürMensch und Küste: am Beispiel der Nordseeinsel Sylt.Springer, Berlin, Heidelberg, New York.

Davie, M and Buffett, B (2001) ‘A numerical model for the for-mation of gas hydrate below the seafloor’. Journal of Geo-physical Research 106(B1): pp185–99.

del Ninno, C, Dorosh, P A, Smith, L C and Roy, D K (2001) The1998 Floods in Bangladesh: Disaster Impacts, Household Cop-ing Strategies, and Response. Research Report 122. Interna-tional Food Policy Research Insitute (IFPRI), Washington,DC.

Deutsche BP (2005) ‘BP plant Kraftwerk mit 90% wenigerEmissionen. Pressemitteilung vom 06. Juli 2005’. DeutscheBP website, http://www.deutschebp.de/genericarticle.do?cat-egoryId=2010149&contentId=7007117 (viewed 3. April2006).

DINAS-COAST Consortium (2004) DIVA 1.0. CD-Rom. PIKPotsdam, Potsdam.

DOE-NETL (US Department of Energy National EnergyTechnology Laboratory) (2005) ‘Program Facts – MethaneHydrates’. NETL website, http://www.netl.doe.gov/scngo(viewed 24. February 2006).

Donner, S D, Skirving, W J, Little, C M, Oppenheimer, M andHoegh-Guldberg, O (2005) ‘Global assessment of coralbleaching and required rates of adaptation under climatechange’. Global Change Biology 11: pp2251–65.

Dowdeswell, J A (2006) ‘Atmospheric science – The GreenlandIce Sheet and global sea-level rise’. Science 311: pp963–4.

Dowsett, H J, Thompson, R S, Barron, J A, Cronin, T M, Ish-man, S E, Poore, R Z, Willard, D A and Holtz jr, T R (1994)‘Joint investigations of the middle pliocene climate I: PRISMpaleoenvironmental reconstructions’. Global and PlanetaryChange 9: pp169-95.

DRM (World Institute for Disaster Risk Management) (2006)‘Disaster Management for Critical Infrastructure’.DRM website, http://www.drmonline.det/projects/infrastruc-ture.htm (viewed 19. January 2006).

8 References

Edenhofer, O (2003) ‘Wege zu einer nachhaltigen Klima- undEnergiepolitik’. Aus Politik und Zeitgeschichte 27: pp18–26.

Edenhofer, O, Held, H and Bauer, N (2005): ‘A regulatoryframework for carbon capturing and sequestration within thePost-Kyoto process’. In Rubin, E S, Keith, D W and Gilboy, CF (eds) Proceedings of the 7th International Conference onGreenhouse Gas Control Technologies. Volume 1: Peer-Reviewed Paper and Plenary Presentations, IEA GreenhouseGas Programme. forthcoming. Edward Elgar, Cheltenham,UK.

Edwards, M and Richardson, J (2004) ‘Impact of climatechange on marine pelagic phenology and trophic mismatch’.Nature 430: pp881–4.

EEA (European Environment Agency) (2005) Vulnerabilityand Adaptation to Climate Change in Europe. EEA, Paris.

EIA (Energy Information Administration) (1998) Natural Gas1998: Issues and Trends. EIA, Washington, DC.

Ellison, J C and Stoddart, D R (1991) ‘Mangrove ecosystemcollapse during predicted sea-level rise: holocene analoguesand implications’. Journal of Coastal Research 7: pp151–65.

El-Raey, M, Dewidar, K R and El-Hattab, M (1999) ‘Adapta-tion to the impacts of sea-level rise in Egypt’. Mitigation andAdaptation Strategies for Global Change 4: pp343-61.

Elsner, W, Otte, C and Yu, I (2005) Klimawandel und präven-tives Risiko- und Küstenschutzmanagement an der deutschenNordseeküste (KRIM):Teilprojekt 5 – Endbericht. Klimawan-del und regionale Wirtschaft.Vermögensschäden und Einkom-mensverluste durch extreme Klimaereignisse sowie Kosten-Nutzen-Analysen von Schutzmaßnahmen. Am Beispiel dernordwestdeutschen Küstenregion. University Bremen, Bre-men.

Emanuel, K (2005) ‘Increasing destructiveness of tropicalcyclones over the past 30 years’. Nature 436: pp686–8.

Emerton, L (1999) Economic Tools for the Management ofMarine Protected Areas in Eastern Africa. Biodiversity Eco-nomics for Eastern Africa. The World Conservation Union(IUCN), Gland.

Engel,A,Thomas, S, Riebesell, U, Rochelle-Newall, E and Zon-dervan, I (2004) ‘Polysaccharide aggregation as a potentialsink of marine dissolved organic carbon’. Nature 428:pp929–32.

Epiney, A (1995) ‘Das „Verbot erheblicher grenzüberschreit-ender Umweltbeeinträchtigungen“: Relikt oderkonkretisierungsfähige Grundnorm?’ Archiv des Völker-rechts 33: pp309ff.

EU Commission (2005) Thematische Strategie für den Schutzund die Erhaltung der Meeresumwelt. Mitteilung der Kommis-sion an den Rat und das Europäische Parlament. EU Docu-ment KOM(2005)504 of 24.10.2005. EU Commission, Brus-sels.

Fairbanks, R G (1989) ‘A 17,000-year glacio-eustatic sea levelrecord: influence of glacial melting rates on the YoungerDryas event and deep-ocean circulation’. Nature 342:pp637–42.

Falkowski, P G, Laws, E A, Baber, R T and Murray, J W (2003):‘Phytoplankton and their role in primary, new, and exportproduction’. In Fasham, M J R (ed) Ocean Biogeochemistry:The Role of the Ocean Carbon Cycle in Global Change.Springer, Berlin, Heidelberg, New York: pp99–121.

Fankhauser, S (1995) Valuing Climate Change: The Economicsof the Greenhouse. Earthscan, London.

Fankhauser, S (1998) The Costs of Adaptation to ClimateChange. GEF Working Paper 16. Global Environment Facil-ity (GEF), Washington, DC.

FAO (Food and Agriculture Organisation) (1995) Code ofConduct for Responsible Fisheries. FAO, Rome.

FAO (Food and Agriculture Organisation) (2000) The State ofWorld Fisheries and Aquaculture 2000. FAO, Rome.

FAO (Food and Agriculture Organisation) (2001) InternationalPlan of Action to Prevent, Deter and Eliminate Illegal, Unre-ported and Unregulated Fishing. FAO, Rome.


FAO (Food and Agriculture Organisation) (2003) FisheriesManagement 2: The Ecosystem Approach to Fisheries. FAOTechnical Report Nr. 4 (Suppl. 2). FAO, Rome.


Fasham, M J R (ed) (2003) Ocean Biogeochemistry:The Role ofthe Ocean Carbon Cycle in Global Change. Springer, Berlin,Heidelberg, New York.

Feely, R, Sabine, C, Lee, K, Berelson, W, Kleypas, J, Fabry, V Jand Millero, F J (2004) ‘Impact of anthropogenic CO2 on theCaCO3 system in the oceans’. Science 305: pp362–6.

Fernando, H J S and McCulley, J L (2005) ‘Coral poachingworsens tsunami destruction in Sri Lanka’. EOS 86: pp301–4.

Fiedler, P.C (2002) ‘Environmental change in the eastern trop-ical Pacific Ocean: review of ENSO and decadal variability’.Marine Ecology Progress Series 244: pp265–83.

Field, D B, Baumgartner, T R., Charles, C D, Ferreira-Bartrina,V and Ohman, M D (2006) ‘Planktonic foraminifera of theCalifornia current reflect 20th-century warming’. Science 311:pp63–6.

Forcada, J, Trathan, P N, Reid, K and Murphy, E J (2005) ‘Theeffects of global climate variability in pup production ofantarctic fur seals’. Ecology 86: pp2408–17.

Frank, K T, Petrie, B, Choi, J S and Leggett,W C (2005) ‘Trophiccascades in a formerly cod-dominated ecosystem’. Science308: pp1621–3.

Freeman, P K, Keen, M and Mani, M (2003) Dealing withIncreased Risk of Natural Disasters: Challenges and Options.International Monetary Fund (IMF), Washington, DC.

Freeth, S J (1987) ‘Life (and death) in Lake Nyos’. Nature 329:pp491.

Freiwald,A, Fosså, J H, Grehan,A, Koslow,T and Roberts, J M(2004) Cold-water Coral Reefs. UNEP-WCMC, Cambridge,New York, UK.

Friends of the Earth (2005) ‘A Citizen's Guide to ClimateRefugees’. Friends of the Earth website, http://www.foe.org.au/download/CitizensGuide.pdf (viewed 15. February2006).

Fujita, R, Bonzon, K, Wilen, J, Solow, A, Arnason, R, Cannon, Jand Polasky, S (2004): ‘Rationality or chaos? Global fisheriesat the crossroads’. In Glover, L K and Earle, S A (eds) Defy-ing Ocean's End. An Agenda for Action. Island Press, Wash-ington, DC: pp139–49.

Gattuso, J-P, Allemande, D and Frankignoulle, M (1999) ‘Pho-tosynthesis and calcification at cellular, organismal and com-munity levels in coral reefs.A review of interactions and con-trol by carbonate chemistry’. American Zoologist 39:pp160–83.

GCIM (Global Commission on International Migration)(2005) Migration in an Interconnected World: New Directionsfor Action. GCIM, Geneva.

GEF (Global Environment Facility) (2005a) Producing Resultsfor the Global Environment.Annual Report 2004. GEF,Wash-ington, DC.

102

103References 8

GEF (Global Environment Facility) (2005b) GEF GlobalAction on Climate Change. GEF Support for Adaptation toClimate Change. GEF, Washington, DC.

Gell, F R and Roberts, C M (2003) ‘Benefits beyond bound-aries: the fishery effects of marine reserves’. Trends in Ecol-ogy and Evolution 18(9): pp448–55.

Gerber, L R, Hyrenbach, K D and Zacharias, M A (2005) ‘Dothe largest protected areas conserve whales or whalers?’ Sci-ence 307: pp525–6.

GESAMP (Joint Group of Experts on the Scientific Aspects ofMarine Environmental Protection) (2001) A Sea of Troubles.GESAMP, The Hague.

Gianni, M and Simpson, W (2005) The Changing Nature ofHigh Seas Fishing: How Flags of Convenience Provide Coverfor Illegal, Unreported and Unregulated Fishing. AustralianDepartment of Agriculture Fisheries and Forestry, Interna-tional Transport Workers’ Federation, WWF International,Sidney, London, Gland.

Gillett, N P, Graf, H F and Osborn, T J (2003): ‘Climate changeand the North Atlantic Oscillation’. In Hurrell, J W, Kushnir,Y, Ottersen, G and Visbeck, M (eds) North Atlantic Oscilla-tion: Climatic Significance and Environmental Impact. Amer-ican Geophysical Union, Washington, DC: pp193–209.

GLOBEC (Global Ocean Ecosystem Dynamics) (1999) Imple-mentation Plan. IGBP Report Nr. 47. GLOBEC Report Nr.13. IGBP Secretariat, Stockholm.

Gloor, M, Gruber, N, Sarmiento, J, Sabine, C L, Feely, R A andRödenbeck, C (2003) ‘A first estimate of present and prein-dustrial air-sea CO2 flux patterns based on ocean interior car-bon measurements and models’. Geophysical Research Let-ters 30(1): pp10-11–10-14.

Gommes, R, du Guerny, J, Nachtergaele, F and Brinkman, R(1998) ‘Potential Impacts of Sea-Level Rise on Populationand Agriculture. FAO SD-Dimensions’. FAO website,http://www.fao.org/sd/eidirect/EIre0045.htm (viewed 22. Feb-ruary 2006).

Gorina-Ysern, M, Gjerde, K and Orbach, M (2004): ‘Oceangovernance: a new ethos through a World Ocean PublicTrust’. In Glover, L K, Earle, S A and Kelleher, G (eds) Defy-ing Ocean's End. An Agenda for Action. Island Press, Wash-ington, DC: pp197–212.

Gornitz, V, Couch, S and Hartig, E K (2002) ‘Impacts of sealevel rise in the New York City metropolitan area’. Globaland Planetary Changes 32: pp61–88.

Greenblatt, J B and Sarmiento, J L (2004): ‘Variability and cli-mate feedback mechanisms in ocean uptake of CO2’. In Field,C B and Raupach, M R (eds) SCOPE 62:The Global CarbonCycle: Integrating Humans, Climate and the Natural World.Island Press, Washington, DC: pp257–275.

Gregg, W W, Conkright, M E, Ginoux, P, O'Reilly, J E andCasey, N W (2003) ‘Ocean primary production and climate:global decadal changes’. Geophysical Research Letters 30:p1809.

Gregory, J M, Huybrechts, P and Raper, S C B (2004) ‘Threat-ened loss of the Greenland ice-sheet’. Nature 428: p616.

Grimsditch, G D and Salm, R V (2005) Coral Reef Resilienceand Resistance to Bleaching. The World Conservation Union(IUCN), Gland.

Gruber, N, Friedlingstein, P, Field, C B, Valentini, R, Heimann,M, Richey, J E, Romero Lankao, P, Schulze, E-D and Chen-Tung,A C (2004): ‘The vulnerability of the carbon cycle in the21st century: an assessment of carbon-climate-human inter-actions’. In Field, C B and Raupach, M R (eds) The GlobalCarbon Cycle. Island Press, Wasington, DC: pp45–77.

Guinotte, J, Buddemeier, R W and Kleypas, J (2003) ‘Futurecoral reef habitat marginality: temporal and spatial effects ofclimate change in the Pacific basin’. Coral Reefs 22: pp551–8.

Hadley Centre (2003) ‘Global Sea Ice Coverage and Sea Sur-face Temperature Data (1856–Present)’. MeteorologicalOffice – Hadley Centre website, http://badc.nerc.ac.uk/data/gisst (viewed 10. February 2006).

Hall, J W, Meadowcroft, I C, Lee, E M and van Gelder, P H A JM (2002) ‘Stochastic simulation of episodic soft coastal cliffrecession’. Coastal Engineering 46: pp159–74.

Hall, J M, Sayers, P B, Walkden, M J A and Panzeri, M (2005)‘Impacts of climate change on coastal flood risk in Englandand Wales: 2030-2100’. Philosophical Transactions of theRoyal Society of London 69: pp343–65.

Hansen, J, Nazarenko, L, Ruedy, R, Sato, M,Willis, J, Del Genio,A, Koch, D, Lacis, A, Lo, K, Menon, S, Novakov, T, Perlwitz, J,Russell, G, Schmidt, G A and Tausnev, N (2005) ‘Earth'senergy imbalance: confirmation and implications’. Science308: pp1431–5.

Hare, S R and Mantua, N J (2000) ‘Empirical evidence forNortheast Pacific regime shifts in 1977 and 1989’. Progress inOceanography 47(2-4): pp99–102.

Haugan, P and Drange, S (1996) ‘Effects of CO2 on the oceanenvironment’. Energy Conversion and Management 37:pp1019–22.

Hilborn, R (2003) ‘Marine reserves: the best option for ouroceans?’ Frontiers in Ecology and the Environment 1(9):pp497–8.

Hoegh-Guldberg, O (1999) ‘Climate change, coral bleachingand the future of the world’s coral reefs’. Marine FreshwaterResources 50: pp839–66.

Hoegh-Guldberg, O (2005) ‘Low coral cover in a high-CO2world’. Journal of Geophysical Research 110: in print.

Hohmann, H (2001): ‘Weltweiter Schutz der Meeresumweltunter besonderer Berücksichtigung des Schutzes von Nord-und Ostsee (einschließlich des Wattenmeeres)’. InGesellschaft für Umweltrecht (ed) Umweltrecht im Wandel.Erich Schmidt, Berlin: pp99–128.

Hoyos, C D, Agudelo, P A, Webster, P J and Curry, J A (2006)‘Deconvolution of the factors contributing to the increase inglobal hurricane intensity’. Science 312: pp94–7.

Hsieh, C-H, Glaser, S M, Lucas, A J and Sugihara, G (2005)‘Distinguishing random environmental fluctuations fromecological catastrophes for the North Pacific Ocean’. Nature435: pp336–40.

Hughes, T P, Baird, A H, Bellwood, D R, Card, M, Connolly, SR, Folke, C, Grosberg, R, Hoegh-Guldberg, O, Jackson, J B C,Kleypas, J, Lough, J, Marshall, P, Nystrom, M, Palumbi, S R,Pandolfi, J M, Rosen, B and Roughgarden, J (2003) ‘Climatechange, human impacts, and the resilience of coral reefs’. Sci-ence 301: pp929–33.

Huisman, J, Nga, N, Thi, P, Karl, D M and Sommeijer, B (2006)‘Reduced mixing generates oscillations and chaos in theoceanic deep chlorophyll maximum’. Nature 439: pp322–5.

Huntley, B, Green, R E, Collingham, Y. C, Hill, J K, Willis, S G,Bartlein, P J, Cramer, W, Hagemeijer, W J M and Thomas, C J(2004) ‘The performance of models relating species geo-graphical distributions to climate is independent of trophiclevel’. Ecology Letters 7: pp417–26.

IEA (International Energy Agency) (2002) CO2 EmissionsFrom Fuel Combustion 1971–2000. IEA, Paris.

IEA (International Energy Agency) (2004) Prospects for CO2Capture and Storage. IEA, Paris.

IEA (International Energy Agency) (2005) Legal Aspects ofStoring CO2. IEA, Paris.

IMBER (Integrated Marine Biogeochemistry and EcosystemResearch) (2005) ‘Science Plan and Implementation Strategy.IGBP Report No. 52’. IGBP Secretariat website,

8 References

http://www.jhu. edu/scor/IMBER/IMBER_SPIS_Final.pdf(viewed 3. February 2006).

IMO (International Maritime Organization) (2004) Report ofthe Twenty-Seventh Meeting of the Scientific Group. IMO,London.

IPCC (Intergovernmental Panel on Climate Change) (1990)Potential Impacts of Climate Change. Report of WorkingGroup II to IPCC. World Meteorological Organization(WMO), United Nations Environment Programme (UNEP),Geneva.

IPCC (Intergovernmental Panel on Climate Change) (2000)Special Report on Emissions Scenarios. Cambridge Univer-sity Press, Cambridge, New York.

IPCC (Intergovernmental Panel on Climate Change) (2001a)Climate Change 2001: The Scientific Basis. Contribution ofWorking Group I to the Third Assessment Report of the IPCC.Cambridge University Press, Cambridge, New York.

IPCC (Intergovernmental Panel on Climate Change) (2001b)Climate Change 2001: Impacts, Adaptation, and Vulnerability.Contribution of Working Group II to the Third AssessmentReport of the IPCC. Cambridge University Press, Cambridge,New York.

IPCC (Intergovernmental Panel on Climate Change) (2005)Special Report on Carbon Dioxide Capture and Storage. Cam-bridge University Press, Cambridge, New York.

Isciences (2003) SRTM30 Enhanced Global Map Data. Eleva-tion/Slope/Aspect. Isciences, Ann Arbor, MI.

IUCN (The World Conservation Union) (1988) Resolution17.38 of the 17th General Assembly of the IUCN. IUCN,Gland, Cambridge.

IUCN (The World Conservation Union) (1994) Guidelines forProtected Areas. IUCN, Gland.

Jackson, J B C, Kirby, M X, Berger, W H, Bjorndal, K A, Bots-ford, L W, Bourque, B J, Bradbury, R H, Cooke, R G, Erland-son, J, Estes, J A, Hughes,T P, Kidwell, S, Lange, C B, Lenihan,H S, Pandolfi, J M, Peterson, C H, Steneck, R S, Tegner, M Jand Warner, R R (2001) ‘Historical overfishing and the recentcollapse of coastal ecosystems’. Science 293: pp629–38.

Jensen, A S (1939) ‘Concerning a change of climate duringrecent decades in the Arctic and Subarctic regions, fromGreenland in the west to Eurasia in the east, and contempo-rary biological and geophysical changes’. Biologiske Med-delelser. Kongelige Danske Videnskabers Selskab XIV: p75.

Jesus, J L (2003): ‘Rocks, new-born islands, sea level rise andmaritime space’. In Frowein, J A, Scharioth, K, Winkelmann,I and Wolfrum, R (eds) Verhandeln für den Frieden. Negotiat-ing for Peace. Liber Amicorum Tono Eitel. Springer, Berlin,Heidelberg, New York: pp579–603.

Jickells, T D, An, Z S, Andersen, K K, Baker, A R, Bergametti,G, Brooks, N, Cao, J J, Boyd, P W, Duce, R A, Hunter, K A,Kawahata, H, Kubilay, N, LaRoche, J, Liss, P S, Mahowald, N,Prospero, J M, Ridgwell, A J, Tegen, I and Torres, R (2005)‘Global iron conncections between desert dust, ocean bio-geochemistry, and climate’. Science 308: pp67–71.

Jimenez, J A and Sanchez-Arcilla,A (1997) ‘Physical impacts ofclimatic change on deltaic coastal systems (II): driving terms’.Climatic Change 35: pp95–118.

Johannessen, O M, Khvorostovsky, K, Miles, M W and Bobylev,L P (2005) ‘Recent ice-sheet growth in the interior of Green-land’. Science 310: pp1013–6.

Johnson, A H (2004) Letter of the “Methane Hydrate AdvisoryCommittee” to the US Department of Energy of 22.11.2004.Methane Hydrate Advisory Committee, Washington, DC.

Joughin, I, Abdalati, W and Fahnestock, M (2004) ‘Large fluc-tuations in speed on Greenland's Jakobshavn Isbrae glacier’.Nature 432: pp608–10.

Kelleher, G (ed) (1999) Guidelines for Marine Protected Areas.Best Practice Protected Area Guidelines Series No. 3. WorldCommission on Protected Areas of IUCN, Gland.

Kellersmann, B (2000) Die gemeinsame, aber differenzierte Ver-antwortlichkeit von Industriestaaten und Entwicklungslän-dern für den Schutz der globalen Umwelt. Springer, Berlin,Heidelberg, New York.

King, J R (2005) Report of the Study Group on Fisheries andEcosystem Responses to Recent Regime Shifts. PICES Scien-tific Report No. 28. North Pacific Marine Science Organiza-tion (PICES), Sidney, BC, Canada.

Klauda, J and Sandler, S I (2005) ‘Global distribution ofmethane hydrate in ocean sediment’. Energy & Fuels 19(2):pp459–70.

Klein, R J T, Nicholls, R J and Mimura, N (1999) ‘Coastal adap-tation to climate change: can the IPCC Technical Guidelinesbe applied?’ Mitigation and Adaptation Strategies for GlobalChange 4(3-4): pp239–52.

Klein, R J T, Nicholls, R J and Thomalla, F (2002): ‘Theresilience of coastal megacities to weather-related hazards: areview’. In Kreimer, A, Arnold, M and Carlin, A (eds) Pro-ceedings of The Future of Disaster Risk: Building Safer Cities.World Bank, Washington, DC: pp111–37.

Kleypas, J, Buddemeier, R H,Archer, D, Gattuso, J-P, Langdon,C and Opdyke, B N (1999) ‘Geochemical consequences ofincreased atmospheric carbon dioxide on coral reefs’. Science284: pp118–20.

Kleypas, J A, Buddemeier, R W, Eakin, C M, Gattuso, J-P,Guinotte, J, Hoegh-Guldberg, O, Iglesias-Prieto, R, Jokiel, PL, Langdon, C, Skirving,W and Strong,A E (2005) ‘Commenton "Coral reef calcification and climate change: the effect ofocean warming"’. Geophysical Research Letters 32:ppL08601.

Kling, G W, Clark, M A, Compton, H R, Devine, J D, Evans, WC, Humphrey,A M, Koenigsberg, E J, Lockwood, J P,Tuttle, ML and Wagner, G N (1987) ‘The 1986 Lake Nyos gas disasterin Cameroon, West Africa’. Science 236: pp169–75.

Klyashtorin, L (2001) Climate change and long-term fluctua-tions of commercial catches. The possibility of forecasting.FAO, Rome.

Knutson, T R and Tuleya, R E (2004) ‘Impact of CO2-inducedwarming on simulated hurricane intensity and precipitation’.Journal of Climatology 17: pp3477–95.

Kropp, J, Eisenack, K and Scheffran, J (2005): ‘Marine overex-ploitation: a syndrome of global change’. In Sonak, S (ed)Multiple Dimensions of Global Environmental Change. TeriPress, New Delhi: in print.

Kullenberg, G (2001) ‘Contributions of marine and coastal arearesearch and observations towards sustainable developmentof large coastal cities’. Ocean & Coastal Management 44:pp283–91.

Kundzewicz, Z W and Schellnhuber, H-J (2004) ‘Floods in theIPCC TAR perspective’. Natural Hazards 31: pp111–28.

Lambeck, K, Anzidei, M, Antonioli, F, Benini, A and Espositol,A (2004) ‘Sea level in Roman times in the central Mediter-ranean and implications for recent change’. Earth and Plane-tary Science Letters 224: pp563–75.

Leatherman, S P (2001) Sea Level Rise: History and Conse-quences. Academic Press, San Diego.

104

105References 8

Lehodey, P, Chai, F and Hampton, J (2003) ‘Modelling climate-related variability of tuna populations from a coupled oceanbiogeochemical-populations dynamics model’. FisheriesOceanography 12: pp483–94.

Levermann, A, Griesel, A, Hofmann, M, Montoya, M andRahmstorf, S (2005) ‘Dynamic sea level changes followingchanges in the thermohaline circulation’. Climate Dynamics24(4): pp347–54.

Levitus, S, Antonov, J and Boyer, T P (2005) ‘Warming of theworld ocean, 1955–2003’. Geophysical Research Letters 32:L02604.

Lindsay, R W and Zhang, J (2005) ‘The thinning of Arctic seaice, 1988-2003: have we passed a tipping point?’ Journal ofClimate 18: pp4879–94.

Linnerooth-Bayer, J and Mechler, R (2005) ‘The potential ben-efits and limitations of micro-insurance as a risk transfermechanism for developing countries. Viewpoint for Interna-tional Day for Disaster Reduction’. MicroInsurance Centrewebsite, http://www.microinsurancecentre.org (viewed 29.November 2005).

Lough, J M and Barnes, D J (2000) ‘Environmental controls ongrowth of the massive coral Porites’. Journal of ExperimentalMarine Biology and Ecology 245: pp225–43.

Lowe, J A, Gregory, J M and Flather, R A (2001) ‘Changes inthe occurrence of storm surges around the United Kingdomunder a future climate scenario using a dynamic storm surgemodel driven by the Hadley Centre climate models’. ClimateDynamics 18: pp179–88.

Lubchenko, J, Palumbi, S R, Gaines, S D and Andelman, S J(2003) ‘Plugging a hole in the ocean: the emerging science ofmarine reserves’. Ecological Applications 13(1): ppS3–S7.

MA (Millennium Ecosystem Assessment) (2005a) Ecosystemsand Human Well-Being: Summary for Decision-Makers.Island Press, Washington, DC.

MA (Millennium Ecosystem Assessment) (2005b): Ecosystemsand Human Well-Being. Volume I: Current State and Trends.Island Press, Washington, DC.

Mai, S, Elsner, A, Meyer, V and Zimmermann, C (2004) Kli-mawandel und präventives Risiko- und Küstenschutzmanage-ment an der deutschen Nordseeküste (KRIM): Teilprojekt 2 –Endbericht. Klimaänderung und Küstenschutz. Franzius Insti-tute, University Hannover, Hannover.

Maslowski, W, Clement, J, Laxon, S and Jakacki, J (2005) ‘Therate of change of the Arctic sea ice cover: 1979–2003 modelresults’. Geophysical Research Abstracts 7: pp10380.

McMichael, A J, Campbell-Lendrum, D, Kovats, S, Edwards, S,Wilkinson, P, Wilson, T, Nicholls, R J, Hales, S, Tanser, F, LeSueur, D, Schlesinger, M and Andronova, N (2004): ‘Climatechange’. In Ezzati, M, Lopez, A D, Rodgers, A and Murray, CJ L (eds) Comparative Quantification of Health Risks: Globaland Regional Burden of Disease due to Selected Major RiskFactors. World Health Organisation (WHO), Geneva:pp1543–649.

McNeil, B I, Matera, R J and Barnes, D J (2004) ‘Coral reef cal-cification and climate change: the effect of ocean warming’.Geophysical Research Letters 31: L22309.

Meier, H E M, Döscher, R and Halkka,A (2004) ‘Simulated dis-tributions of Baltic sea-ice in warming climate and conse-quences for the winter habitat of the Baltic ringed seal’.Ambio 33(4-5): pp249–56.

Meinshausen, M (2006): ‘What does a 2°C target mean forgreenhouse gas concentrations? A brief analysis based onmulti-gas emission pathways and several climate sensitivityuncertainty estimates’. In Schellnhuber, H-J (ed) AvoidingDangerous Climate Change. Cambridge University Press,Cambridge, New York: pp265–79.

Met Office (2006) ‘Catarina Hits Brazil. South Atlantic Hurri-cane Breaks all the Rules’. Met Office website, http://www.metoffice.com/sec2/sec2cyclone/catarina.html (viewed14. February 2006).

Methane Hydrate Advisory Committee (2002) ‘Report of theMethane Hydrate Advisory Committee on Methane HydrateIssues and Opportunities – Including Assessment of Uncer-tainty of the Impact of Methane Hydrate on Global ClimateChange, Dezember 2002’. U.S. Department of Energy (DOE)website, http://www.fe.doe.gov/programs/oilgas/hydrates/Methane_Hydrates_Advisory_Committee.html (viewed 24.February 2006).

MH21 (Research Consortium for Methane Hydrate Resourcesin Japan) (2005) ‘Background and Organization’. MH 21website, http://www.mh21japan.gr.jp/english/mh21-2. html(viewed 15. February 2006).

Mignot, J A, Ganopolski, A and Levermann, A (submitted)‘Atlantic subsurface temperature response to a shut-down ofthe overturning circulation’. Journal of Climate.

Milazzo, M (1998) Subsidies in World Fisheries – A Reexamina-tion. World Bank Technical Paper 406. Fisheries Series. WorldBank, Washington, DC.

Milkov, A V (2004) ‘Global estimates of hydrate-bound gas inmarine sediments: how much is really out there?’ Earth-Sci-ence Reviews 66(3-4): pp183–97.

Mirza, M M Q, Warrick, R A and Ericksen, N J (2003) ‘Theimplications of climate change on floods of the Ganges,Brahmaputra and Meghna rivers in Bangladesh’. ClimaticChange 57: pp287–318.

Monnett, C, Gleason, J and Rotterman, L (2005) Report at the16th Biennial Conference on the Biology of Marine Mammals.Conference Secretariat, San Diego, USA.

Montaggioni, L F (2005) ‘History of Indo-Pacific coral reef sys-tems since the last glaciation: Development patterns and con-trolling factors’. Earth-Science Reviews 71: pp1–75.

Morling, P (2004) ‘The economics of marine protected areas inthe High Seas’. The George Wright Forum 21(4): pp49–61.

MPI for Meteorology (2005) Klimaänderung so rasant wienoch nie. Press Release No. B62/C29/2005(146) of 29.9.2005.MPI for Meteorology, Hamburg.

Mumby, P J, Dahlgren, C P, Harborne, A R, Kappel, C V,Micheli, F, Brumbaugh, D R, Holmes, C E, Mendes, J M,Broad, K, Sanchirico, J N, Buch, K, Box, S, Stoffle, R W andGill, A B (2006) ‘Fishing, trophic cascades, and the process ofgrazing on coral reefs’. Science 311: pp98–101.

Murray, S N,Ambrose, R F, Bohnsack, J A, Botsford, L W, Carr,M H, Davis, G E, Dayton, P K, Gotshall, D, Gunderson, D R,Hixon, M A, Lubchenco, J, Mangel, M, MacCall, A, McArdle,D A, Ogden, J C, Roughgarden, J, Starr, R M,Tegner, M J andYoklavich, M M (1999) ‘No-take reserve networks: sustainingfisheries populations and marine ecosystems’. Fisheries 24(1):pp11–25.

Münchener Rück (1997) ‘Überschwemmung und Ver-sicherung’. Münchener Rück website, http://www.munichre.com (viewed 16. March 2006).

Münchener Rück (2005a) Wetterkatastrophen und Klimawan-del. Sind wir noch zu retten? Münchener Rück, Munich.

Münchener Rück (2005b) ‘Mikroversicherung wichtige Ein-richtung zur Armutsbekämpfung’. Münchener Rück Stiftungwebsite, http://www.epo.de (viewed 29. November 2005).

NASA (National Aeronautics and Space Administration)(2005) ‘Arctic Sea Ice Continues to Decline, Arctic Tempera-tures Continue to Rise in 2005’. NASA Headquarters web-site, http://www.nasa.gov/centers/goddard/news/topstory/2005/arcticice_decline.html (viewed 13. March 2006).

8 References

Nicholls, R J (1995) ‘Coastal megacities and climate change’.GeoJournal 37(3): pp369–79.

Nicholls, J R (2003) Case Study on Sea-Level Rise Impacts.OECD Workshop on the Benefits of Cliamte Policy: Improv-ing Information for Policy Makers. DocumentENV/EPOC/GSP(2003)9/FINAL. Organization for Eco-nomic Co-operation and Development (OECD), Paris.

Nicholls, R J (2004) ‘Coastal flooding and wetland loss in the21st century: changes under the SRES climate and socio-eco-nomic scenarios’. Global Environmental Change 14: pp69–86.

Nicholls, R J, Tol, R S J and Vafeidis, A T (2005) Global Esti-mates of the Impact of a Collapse of the West Antarctic IceSheet: An Application of FUND. Working Paper FNU-78.University Hamburg, Center for Marine and AtmosphericSciences, Hamburg.

Nicholls, R J, Townend, I and Cooper, N (2006) ‘Coastal engi-neering and management’. In Thorne, C R, Evans, E P andPenning-Rowsell, E C (eds) Future Flood and Coastal Ero-sion Risks in the UK. Thomas Telford, London.

NOAA-ESRL (Global Monitoring Division) (2006) ‘Interac-tive Atmospheric Data Visualization. Mauna Loa, Hawaii,United States’. NOAA ESRL website, http://www.cmdl.noaa.gov/ccgg/iadv/ (viewed 20. March 2006).

NODC (National Oceanographic Data Center) (2001) WorldOcean Database 2001. NODC, Silver Spring, MD.

Nosengo, N (2003) ‘Save our city!’ Nature 424: pp608–9.

NRC (National Research Council) (2001) Marine ProtectedAreas: Tools for Sustaining Ocean Ecosystem. Committee onthe Evaluation, Design, and Monitoring of Marine Reservesand Protected Areas in the United States. Ocean StudiesBoard, National Research Council. NRC, Washington, DC.

NSIDC (National Snow and Ice Data Center) (2002) ‘Larsen BIce Shelf Collapses in Antarctica’. NSIDC website,http://nsidc.org/iceshelves/larsenb2002/ (viewed 24. January2006).

OECD (Organisation for Economic Co-operation and Devel-opment) (2002): Aid Targeting the Objectives of the Rio Con-ventions 1998–2000. A Contribution by the DAC Secretariatfor the Information of Participants at the World Summit forSustainable Development in Johannesburg in August 2002.Paris, OECD.

Oppenheimer, M and Alley, R B (2004) ‘The West Antarctic icesheet and long term climate policy – An editorial comment’.Climatic Change 64(1-2): pp1–10.

Orr, J C, Fabry,V J,Aumont, O, Bopp, L, Doney, S C, Feely, R A,Gnanadesikan, A, Gruber, N, Ishida, A, Joos, F, Key, R M,Lindsay, K, Maier-Reimer, E, Matear, R, Monfray, P,Mouchet,A, Najjar, R G, Plattner, G-K, Rodgers, K B, Sabine,C L, Sarmiento, J L, Schlitzer, R, Slater, R D, Totterdell, I J,Weirig, M-F,Yamanaka,Y and Yool,A (2005) ‘Anthropogenicocean acidification over the twenty-first century and itsimpact on calcifying organisms’. Nature 437: pp681–6.

OSPAR (Oslo Paris Commission) (2003) Joint MinisterialMeeting of the Helsinki and OSPAR Commissions (JMM),Bremen 25-26 June 2003. Declaration of the Joint MinisterialMeeting of the Helsinki and OSPAR Commissions. OSPAR,Bremen.

Palumbi, S R (2003) Marine Reserves: A Tool for EcosystemManagement and Conservation. Pew Oceans Commission,Washington, DC.

Pandolfi, J M, Bradbury, R H, Sala, E, Hughes, T P, Bjorndal, KA, Cooke, R G, McArdle, D, McClenachan, L, Newman, M JH, Paredes, G,Warner, R R and Jackson, J B C (2003) ‘Globaltrajectories of the long-term decline of coral reef ecosystems’.Science 301: pp955–8.

Parker, D E, Folland, C K and Jackson, M (1995) ‘Marine sur-face temperature: observed variations and data require-ments’. Climate Change 31: pp559–600.

Parmesan, C and Yohe, G (2003) ‘A globally coherent finger-print of climate change impacts across natural systems’.Nature 421: pp37–42.

Patz, J A, Campbell-Lendrum, D H, Holloway,T and Foley, J A(2005) ‘Impact of regional climate change on human health’.Nature 438: pp310–7.

Pauly, D, Christensen, V., Guenette, S, Pitcher, T J, Sumaila, UR, Walters, C J, Watson, R and Zeller, D (2002) ‘Towards sus-tainability in world fisheries’. Nature 418: pp689–95.

Peltier, W R (2004) ‘Global glacial isostasy and the surface ofthe ice-age earth: The ice-5G (VM2) model and grace’.Annual Review Of Earth And Planetary Sciences 32:pp111–49.

Perry,A L, Low, P J, Ellis, J R and Reynolds, J D (2005) ‘Climatechange and distribution shifts in marine fishes’. Science 308:pp1912–5.

PICES (North Pacific Marine Science Organization) (2004)Marine Ecosystems of the North Pacific. PICES ScientificReport No. 1. PICES, Sidney, Canada.

Pikitch, E K, Santora, C, Babcock, E A, Bakun, A, Bonfil, R,Conover, D O, Dayton, P K, Doukakis, P, Fluharty, D L, Hen-eman, B, Houde, E D, Link, J, Livingston, P A, Mangel, M,McAllister, M K, Pope, J and Sainsbury, K J (2004) ‘Ecosys-tem-based fishery management’. Science 305: pp346–7.

Platzöder, R. (2001): ‘The United Nations Convention on theLaw of the Sea and Marine Protected Areas on the HighSeas’. In Thiel, H and Koslow, J A (eds) Managing Risks toBiodiversity and the Environment on the High Sea, IncludingTools Such as Marine Protected Areas – Scientific Require-ments and Legal Aspects. German Federal Agency for NatureConservation, Bonn: pp137–42.

Ploetz, C (2003) ‘Sequestrierung von CO2: Technologien,Potenziale, Kosten und Umweltauswirkungen’. Expertise forthe WBGU Report ‘World in Transition:Towards SustainableEnergy Systems’. WBGU website, http://www.wbgu. de/wbgu_jg2003_ex07.pdf.

Polovina, J J (2005) ‘Climate variation, regime shifts, and impli-cations for sustainable fisheries’. Bulletin of Marine Science76(2): pp233–44.

Pörtner, O H, Langenbuch, M and Reipschläger,A (2004) ‘Bio-logical impact of elevated ocean CO2 concentrations: lessonsfrom animal physiology and earth history’. Journal ofOceanography 60: pp705–18.

Pörtner, H O (2005) ‘Auswirkungen von CO2-Eintrag und Tem-peraturerhöhung auf die marine Biosphäre’. Expertise forWBGU Special Report ‘The Future Oceans – Warming Up,Rising High, Turning Sour’. WBGU website, http://www.wbgu.de/wbgu_sn2005_ex04.pdf.

Prentice, I C, Farquhar, G D, Fasham, M J R, Goulden, M L,Heimann, M, Jaramillo, V J, Kheshgi, H S, Le Quéré, C,Scholes, R J and Wallace, D W R (2001): ‘The carbon cycleand atmospheric carbon dioxide’. In IntergovernmentalPanel on Climate Change (IPCC) (ed) Climate Change 2001:The Scientific Basis. Contribution of Working Group I to theThird Assessment Report of the IPCC. Cambridge UniversityPress, Cambridge, New York: pp183–239.

Proelß, A (2004) Meeresschutz im Völker- und Europarecht.Das Beispiel des Nordostatlantiks. Duncker & Humblot,Berlin.

Rahmstorf, S (2000) ‘The thermohaline ocean circulation: Asystem with dangerous thresholds? An Editorial Comment’.Climatic Change 46: pp247–56.

106

107References 8

Rahmstorf, S (2002) ‘Ocean circulation and climate during thepast 120,000 years’. Nature 419: pp207–14.

Raper, S C B and Braithwaite, R J (2006) ‘Low sea level riseprojections from mountain glaciers and icecaps under globalwarming’. Nature 439: pp311–3.

Raven, J, Caldeira, K, Elderfield, H, Hoegh-Guldberg, O, Liss,P S, Riebesell, U, Shepherd, J,Turley, C and Watson,A J (2005)Ocean Acidification Due to Increasing Atmospheric CarbonDioxide. Policy Document 12/05. The Royal Society, London.

Ray, P K (2005) ‘Quantitative assessment of marine gashydrates underway at mms’.The National Energy TechnologyLaboratory Methane Hydrate Newsletter (winter): pp5–7.

Reaka-Kudla, M L (1997): ‘The global biodiversity of coralreefs: A comparison with rainforests’. In Reaka-Kudla, M L,Wilson, D E and Wilson, E O (eds) Biodiversity II: Under-standing and Protecting Our Natural Resources. JosephHenry/National Academy Press, Washington, DC: pp83–108.

Reid, P C, Holliday, N P and Smyth, T J (2001) ‘Pulses in theeastern margin current and warmer water off the north westEuropean shelf linked to North Sea ecosystem changes’.Marine Ecology Progress Series 215: pp283–7.

Richardson,A J and Schoeman, D S (2004) ‘Climate impact onplankton ecosystems in the Northeast Atlantic’. Science 305:pp1609–12.

Riebesell, U, Zondervan, I, Rost, B,Tortell, P D, Zeebe, R E andMorel, F M M (2000) ‘Reduced calcification of marine plank-ton in response to increased atmospheric CO2’. Nature 407:pp364–7.

Riebesell, U (2004) ‘Effects of CO2 enrichment on marine phy-toplankton’. Journal of Oceanography 60: pp719–29.

Rignot, E and Jacobs, S (2002) ‘Rapid bottom melting wide-spread near Antarctic ice sheet grounding lines’. Science 296:pp2020–3.

Rignot, E, Vaughan, D G, Schmeltz, M, Dupont, T andMacAyeal, D (2002) ‘Acceleration of Pine Island andThwaites Glaciers, West Antarctica’. Annals Of Glaciology34: pp189–94.

Rignot, E, Casassa, G, Gogineni, P, Krabill, W, Rivera, A andThomas, R (2004) ‘Accelerated ice discharge from the Antarc-tic Peninsula following the collapse of Larsen B ice shelf’.Geophysical Research Letters 31(18): L18401.

Rignot, E and Kanagaratnam, P (2006) ‘Changes in the veloc-ity structure of the Greenland ice sheet’. Science 311:pp986–90.

Rodrigues, A S L, Andelman, S J, Bakarr, M I, Boitani, L,Brooks,T M, Cowling, R M, Fishpool, L D C, da Fonseca, G AB, Gaston, K J, Hoffman, M, Long, J S, Marquet, P A, Pilgrim,J D, Pressey, R L, Schipper, J, Sechrest, W, Stuart, S N, Under-hill, L G, Waller, R W, Watts, M E J and Yan, X (2004) ‘Effec-tiveness of the global protected area network in representingspecies diversity’. Nature 428: pp641–3.

Rogner, H-H (1997) ‘An assessment of world hydrocarbonresources’. Annual Review of Energy and the Environment22: pp217–62.

Rosenzweig, C and Solecki, W D (eds) (2001) Climate Changeand a Global City: The Potential Consequences of ClimateVariability and Change. Metro East Coast. Report for the USGlobal Change Research Program. National Assessment of thePotential Consequences of Climate Variability and Change forthe United States. Columbia Earth Institute, New York.

Rost, B and Sültemeyer, D (2003) ‘Carbon acquisition ofbloom-forming marine phytoplankton’. Limnology andOceanography 48(1): pp55–67.

Rothrock, D D,Yu,Y. and Maykut, G A (1999) ‘Thinning of theArctic sea ice cover’. Geophysical Research Letters 26:pp3469–72.

Rowan, B (2004) ‘Thermal adaptation in reef coral symbionts’.Nature 430: pp742.

Royal Commission on Environmental Pollution (2004) Turningthe Tide: Addressing the Impact of Fisheries on the MarineEnvironment. Twenty-fifth Report. Royal Commission onEnvironmental Pollution, London.

Ruddiman, W F (2000) Earth’s climate: past and future. Free-man, New York.

Sabine, C S, Heimann, M, Artaxo, P, Bakker, C-T, Chen, A,Field, C B, Gruber, N, LeQuere, C, Prinn, R G, Richey, J E,Romero-Lankao, Sathaye, P J and Valentini, R (2003): ‘Cur-rent status and past trends of the carbon cycle’. In Field, C Band Raupach, M R (eds) Toward CO2 Stabilization: Issues,Strategies, and Consequences. Island Press, Washington, DC:pp17-44.

Sabine, C L, Feely, R A, Gruber, N, Key, R M, Lee, K, Bullister,J L, Wanninkhof, R, Wong, C S, Wallace, D W R, Tilbrook, B,Millero, F J, Peng,T-H, Kozyr,A, Ono,T and Rios,A F (2004)‘The oceanic sink for anthropogenic CO2’. Science 305:pp367–71.

Sands, P (2003) Principles of International Environmental Law.Second Edition. Cambridge University Press, Cambridge,New York.

Sarmiento, J L, Gruber, N, Brzezinski, M A and Dunne, J P(2003) ‘High-latitude controls of thermocline nutrients andlow latitude biological productivity’. Nature 427: pp56–60.

Sarmiento, J L, Slater, R, Barber, R, Bopp, L, Doney, S C, Hirst,A C, Kleypas, J, Matear, R, Mikolajewicz, U, Monfray, P,Soldatov, V, Spall, S A and Stouffer, R J (2004) ‘Response ofocean ecosystems to climate warming’. Global Biogeochemi-cal Cycles 18: ppGB3003.

Sauter, E J, Muyakshin, S I, Charlou, J-L, Schlüter, M, Boetius,A, Jerosch, K, Damm, E, Foucher, J-P and Klages, M (2006)‘Methane discharge from a deep-sea submarine mud volcanointo the upper water column by gas hydrate-coated methanebubbles’. Earth and Planetary Science Letters 243(3-4):pp354–65.

Scambos, T A, Bohlander, J A, Shuman, C A and Skvarca, P(2004) ‘Glacier acceleration and thinning after ice shelf col-lapse in the Larsen B embayment, Antarctica’. GeophysicalResearch Letters 31(18): ppL18402.

Schallenberg, M, Hall, C J and Burns, C W (2003) ‘Conse-quences of climate-induced salinity increases on zooplanktonabundance and diversity in coastal lakes’. Marine EcologyProgress Series 251: pp181–9.

Scharroo, R, Smith, W H F and Lillibridge, J F (2005) ‘Satellitealtimetry and the intensification of Hurricane Katrina’. Eos86: pp366–7.

Scheffer, M, Carpenter, S, Foley, J A, Folkes, C and Walker, B H(2001) ‘Catastrophic shifts in ecosystems’. Nature 413:pp591–6.

Schelling, T C (1992) ‘Some economics of global warming’.American Economic Review 82(1): pp1–14.

Schellnhuber, H J, Cramer, W, Nakicenovic, N, Wigley, T andYohe, G (2006) Avoiding Dangerous Climate Change. Cam-bridge University Press, Cambridge, New York.

Schindler, J and Zittel, W (2000a) Fossile Energiereserven (nurErdöl und Erdgas) und mögliche Versorgungsengpässe ausEuropäischer Perspektive. Endbericht. LB-SystemtechnikGmbH, Ottobrunn.

Schindler, J and Zittel, W (2000b) Öffentliche Anhörung vonSachverständigen (17.10.2000) durch die Enquête-Kommis-sion des Deutschen Bundestages „Nachhaltige Energiever-sorgung unter den Bedingungen der Globalisierung und derLiberalisierung“ zum Thema „Weltweite Entwicklung derEnergienachfrage und der Ressourcenverfügbarkeit“.

8 References

Schriftliche Stellungnahme zu ausgewählten Fragen der Kom-mission. LB-Systemtechnik GmbH, Ottobrunn.

Schlesinger, W H (1997) Biogeochemistry. An Analysis ofGlobal Change. Academic Press, San Diego, London, Boston.

Schmittner, A (2005) ‘Decline of the marine ecosystem causedby a reduction in the Atlantic overturning circulation’. Nature434: pp628–33.

Schwartz, P and Randall, D (2003) ‘An Abrupt Climate ChangeScenario and Its Implications for United States NationalSecurity.A report commissioned by the U.S. Defense Depart-ment’. EMS website, http://www.ems.org/climate/pentagon_climatechange.pdf (viewed 13. February 2006).

Sea Wind Europe (2003) ‘A Realistic Vision for the Develop-ment of Offshore Wind Power to 2020. Studie von GarradHassan und anderen im Auftrag von Greenpeace Interna-tional’. Greenpeace International website, http://www.green-peace.org/raw/content/usa/press/reports/ sea-wind-europe.pdf (viewed 8. March 2006).

SEEDS (Sustainable Environment & Ecological DevelopmentSociety) (2005) Strengthening the Resilience of Local Com-munity in Coastal Areas to Water Related Natural Disasters.Integrating Risk Reduction in Recovery. SEEDS, New Delhi.

Shah,T, Molden, D, Sakthivadivel, R and Seckler, D (2000) TheGlobal Groundwater Situation: Overview of Opportunitiesand Challenges. International Water Management Institute,Colombo, Sri Lanka.

Sherif, M M and Singh,V P (1999) ‘Effect of climate change onsea water intrusion in coastal aquifers’. HydrologicalProcesses 13: pp1277–87.

Shirayama, Y. and Thornton, H (2005) ‘Effect of increasedatmospheric CO2 on shallow water marine benthos’. Journalof Geophysical Research 110: ppC09S8.

Siddique,A K, Baqui,A H, Eusof,A and Zaman, K (1991) ‘1988floods in Bangladesh - pattern of illness and causes of death’.Journal of Diarrhoeal Diseases Research 9(4): pp310–4.

Siegfriedsen, S, Lehnhoff, M and Prehn,A (2003) ‘Primary mar-kets for offshore wind energy outside the European Union’.Wind Engineering 27(5): pp419–30.

Skjoldal, H R (ed) (2004) The Norwegian Sea Ecosystem. TapirAcademic Press, Trondheim.

Skjoldal, H R and Sætre, R (2004):‘Climate and ecosystem vari-ability’. In Skjoldal, H R (ed) The Norwegian Sea Ecosystem.Tapir Academic Press, Trondheim: pp507–34.

Slade, N T (2001)‘Environmental concerns of small islands’.Environmental Policy and Law 31(3): 157–8.

Small, C and Nicholls, R J (2003) ‘A global analysis of humansettlement in coastal zones’. Journal of Coastal Research 19:pp584–99.

Smetacek, V and Nicol, S (2005) ‘Polar ocean ecosystems in anchanging world’. Nature 437: pp362–8.

Smith, J B and Lazo, J K (2001) ‘A summary of climate changeimpact assessments from the US Country Studies Pro-gramme’. Climatic Change 50: pp1–29.

Smith, D E, Shi, S, Brooks, C L, Cullingford, R A, Dawson,A G,Dawson, S, Firth, C R, Foster, I D L, Fretwell, P T, Haggart, BA, Holloway, L K and Long, D (2004) ‘The holocene storeggaslide Tsunami in the United Kingdom’. Quaternary ScienceReviews 23(23-24): pp2291–321.

Snedaker, S, Meeder, J, Ross, M and Ford, R (1994) ‘Discussionof Ellison, Joanna C and Stoddart, David R, 1991: Mangroveecosystem collapse during predicted sea-level rise: holoceneanalogues and implications’. Journal of Coastal Research10(2): pp497–8.

SOLAS (Surface-Ocean Lower-Atmosphere Study) (2004)Science Plan and Implementation Strategy. IGBP Report 50.IGBP Secretariat, Stockholm.

Soto, C G (2002) ‘The potential impacts of global climatechange on marine protected areas’. Reviews in Fish Biologyand Fisheries 11: pp181–95.

SRU (Sachverständigenrat für Umweltfragen) (2004) Meeres-umweltschutz für Nord- und Ostsee. Sondergutachten. SRU,Berlin.

Stanley, D J and Warne,A G (1998) ‘Nile delta in its destructionphase’. Journal of Coastal Research 14(3): pp794–825.

Statoil (2005) CO2 Capture and Underground Storage. Presen-tation of Mr. Olav Kaarstad, held for WBGU in Stavanger,Norway, on 12.10.2005. Statoil, Stavanger.

Steele, J H (1998) ‘Regime shifts in marine ecosystems’. Eco-logical Applications 8(1 supplement): ppS33–S6.

Steffen, W, Sanderson, A, Tyson, P D, Jäger, J, Matson, P A,Moore III, B, Oldfield, F, Richardson, K, Schellnhuber, H-J,Turner II, B L and Wasson, R J (2004) Global Change and theEarth System.A Planet Under Pressure. Springer, Berlin, Hei-delberg, New York.

Steffen, K and Huff, R (2005) ‘Greenland Melt Extent 2005’.Cooperative Institute for Research in Environmental Sci-ences (CIRES) website, http://cires.colorado.edu/ science/groups/steffen/greenland/melt2005/ (viewed 8. March 2006).

Stive, M J F (2004) ‘How important is global warming forcoastal erosion?’ Climatic Change 64: pp27–39.

Stommel, H (1961) ‘Thermohaline convection with two stableregimes of flow’. Tellus 13: pp224–30.

Stone, C D (2004) ‘Common but differentiated responsibilitiesin international law’. American Journal of International Law98: pp276ff.

Stone, D A and Allen, M R (2005) ‘The end-to-end attributionproblem: from emissions to impacts’. Climatic Change 71:pp303–18.

Strong, A E, Kearns, E J and Gjovig, K K (2000) ‘Sea surfacetemperature signals from satellites: An update’. GeophysicalResearch Letters 27: pp1667–70.

SwissRe (Schweizerische Rückversicherungs-Gesellschaft)(1998) ‘Überschwemmungen: Ein versicherbares Risiko?’SwissRe website, http://www.swissre.com/INTERNET/pws-filpr.nsf/vwFilebyIDKEYLu/WWIN-4 V2BKW/$FILE /floods.Paras.0002.File.pdf (viewed 8. March 2006).

Tamburri, M N, Peltzer, E T, Friederich, G E,Aya, I,Yamane, Kand Brewer, P G (2000) ‘A field study of the effects of CO2ocean disposal on mobile deep-sea animals’. Marine Chem-istry 72: pp95–101.

Titus, J G (1990) ‘Greenhouse effect, sea level rise, and landuse’. Land Use Policy 7(2): pp138–53.

Titus, J G, Park, R A, Leatherman, S P, Weggel, J R, Greene, MS, Mausel, P W, Brown, S, Gaunt, G, Trehan, M and Yohe, G(1991) ‘Greenhouse effect and sea level rise: the cost of hold-ing back the sea’. Coastal Management 19: pp171–204.

Turley, C, Blackford, J C, Widdicombe, S, Lowe, D, Nightingale,P D and Rees,A P (2006): ‘Reviewing the impact of increasedatmospheric CO2 on oceanic pH and the marine ecosystem’.In Schellnhuber, H J, Cramer, W, Nakicenovic, N, Wigley, Tand Yohe, G (eds) Avoiding Dangerous Climate Change.Cambridge University Press, Cambridge, New York: pp65–70.

Turner, R K, Subak, S and Adger,W N (1996) ‘Pressures, trends,and impacts in coastal zones: interactions between socioeco-nomic and natural systems’. Environmental Management20(2): pp159–73.

Tyler, P A (2003) ‘Disposal in the deep sea: analogue of natureor faux ami?’ Environmental Conservation 30: pp26–39.

Tyrell, T, Holligan, P M and Mobley, C D (1999) ‘Opticalimpacts of oceanic coccolithophore blooms’. Journal of Geo-physical Research 104(C2): pp3223–41.

108

109References 8

UN (United Nations) (ed) (2004) Human Settlements on theCoast. UN Atlas of the Oceans. UN, New York.

UNEP (United Nations Environment Programme) (2002)GEO – Global Environment Outlook 3. Past, Present andFuture Perspectives. London: Earthscan.

UNEP (United Nations Environment Programme) (2005)Pacific Island Villagers First Climate Change „Refugees“.UNEP News Release. UNEP, Nairobi.

UNESCO (United Nations Educational Scientific and CulturalOrganization) (1972) Übereinkommen zum Schutz des Kul-tur- und Naturerbes der Welt. Organisation der VereintenNationen für Bildung,Wissenschaft, Kultur und Kommunika-tion. Deutsche Übersetzung aus dem Bundesgesetzblatt,Jahrgang 1977, Teil II, Nr. 10. UNESCO website,http://www.unesco.de/c_bibliothek/info-welterbekonvention.pdf (viewed 8. March 2006).

UNESCO (United Nations Educational Scientific and CulturalOrganization) (2006) World Heritage List. Results by Coun-try. UNESCO website, http://whc.unesco.org/en/list/ (viewed27. March 2006).

UNFCCC (United Nations Framework Convention on Cli-mate Change) (1992) Framework Convention on ClimateChange. United Nations (UN), New York.

UNGA (United Nations General Assembly) (2004) Oceansand the Law of the Sea. UNGA Resolution 59/24. UnitedNations (UN), New York.

UN ISDR (United Nations Interagency Secretariat of theInternational Strategy for Disaster Reduction) (2005a) Inter-national Strategy for Disaster Reduction. Draft ResolutionSubmitted by the Vice-Chairman of the Committee, Mr. Ste-fano Toscano (Switzerland), on the Basis of Informal Consul-tations Held on Draft Resolution A/C.2/60/L.25. UN ISDR,Geneva.

UN ISDR (United Nations Interagency Secretariat of theInternational Strategy for Disaster Reduction) (2005b) Nat-ural Disasters and Vulnerability. Draft Resolution Submittedby the Vice-Chairman of the Committee, Mr. Stefano Toscano(Switzerland), on the Basis of Informal Consultations Held onDraft Resolution A/C.2/60/L.27. UN ISDR, Geneva.

UN ISDR (United Nations Interagency Secretariat of theInternational Strategy for Disaster Reduction) (2005c)Progress Report:Worldwide Survey of Early Warning Systems.A Briefing Prepared for the EOSG by the ISDR Secretariat, 31July 2005. UN ISDR, Geneva.

Velicogna, I and Wahr, J (2006) ‘Measurements of time-vari-able gravity show mass loss in Antarctica’. Science 311:pp1754–6.

Veron, J E N (1986) Corals of Austrakia and the Indo-Pacific.Angus and Robertson, London, Sidney.

Virdin, J and Schorr, J (2001) Hard Facts, Hidden Problems: AReview of current Data on Fishing Subsidies. World WildlifeFund, Brussels.

Waelbroeck, C, Labeyrie, L, Michel, E, Duplessy, J C,McManus, J F, Lambeck, K, Balbon, E and Labracherie, M(2002) ‘Sea-level and deep water temperature changesderived from benthonic foraminifera isotopic records’. Qua-ternary Science Reviews 21: pp295–305.

Walsh, N P (2003) ‘St Petersburg races to halt floods’. TheGuardian (29.07.2003), St. Petersburg.

Warner, R (2001): ‘Marine protected areas beyond nationaljurisdiction: existing legal principles and a future interna-tional framework’. In Haward, M (ed) Integrated OceansManagement: Issues in Implementing Australia’s Oceans Pol-icy. Cooperative Research Centre for Antarctica and theSouthern Ocean, Hobart: pp55–76.

Watson, A J and Orr, J C (2003): ‘Carbon dioxide fluxes in theglobal ocean’. In Fasham, M J R (ed) Ocean Biogeochemistry.The IGBP Series. Springer, Berlin, Heidelberg, New York.

Wavenet (2003) ‘Results from the Work of the European The-matic Network on Wave Energy’. Wavenet (European WaveEnergy Thematic Network) website, http://www.wave-energy.net/Library/WaveNet_FullReport(11.1).pdf (viewed27. May 2005).

WBGU (German Advisory Council on Global Change) (1995)Scenario for the Derivation of Global CO2 Reduction Targetsand Implementation Strategies. Statement on the Occasion ofthe First Conference of the Parties to the Framework Conven-tion on Climate Change in Berlin. 1995 Special Report.WBGU, Bremerhaven.

WBGU (German Advisory Council on Global Change) (2001)World in Transition: Conservation and Sustainable Use of theBiosphere. 1999 Report. Earthscan, London.

WBGU (German Advisory Council on Global Change) (2002)Charging the Use of Global Commons. Special Report 2002.WBGU, Berlin.

WBGU (German Advisory Council on Global Change) (2003)Climate Protection Strategies for the 21st Century: Kyoto andbeyond. Special Report 2003. WBGU, Berlin.

WBGU (German Advisory Council on Global Change) (2004)World in Transition: Towards Sustainable Energy Systems.2003 Report. Earthscan, London.

WCDR (World Conference on Disaster Reduction) (2005)Hyogo Framework for Action 2005–2015: Building theResilience of Nations and Communities to Disasters. ExtractFrom the Final Report of the World Conference on DisasterReduction (Document A/CONF.206/6). International Strat-egy for Disaster Reduction (ISDR), Kyoto.

Webster, P J, Holland, G J, Curry, J A and Chang, H-R (2005)‘Changes in tropical cyclone number, duration, and intensityin a warming environment’. Science 309: pp1844–6.

WHO (World Health Organisation) (2002) Flooding: HealthEffects and Preventive Measures. Fact Sheet 05/02. WHO,Copenhagen, Rome.

Wilkinson, C (2004) Status of Coral Reefs of the World 2004.Australian Institute of Marine Sciences, Townsville.

Willis, J K, Roemmich, D and Cornuelle, B (2004) ‘Interannualvariability in upper ocean heat content, temperature, andthermosteric expansion on global scales’. Journal of Geo-physical Research-Oceans 109: ppC12036.

Winguth, A, Mikolajewicz, U, Gröger, M, Maier-Reimer, E,Schurgers, G and Vizcaino, M (2005) ‘Centennial-scale inter-actions between the carbon cycle and anthropogenic climatechange using a dynamic Earth system model’. GeophysicalResearch Letters 32: ppdoi:10.1029/2005GL023681.

Wolfrum, R (2000): ‘International environmental law: pur-poses, principles and means of ensuring compliance’. In Mor-rison, F L and Wolfrum, R (eds) International, Regional andNational Environmental Law. Kluwer Law International,TheHague, London, Boston: pp3–70.

Wood, L J, Fish, L, Laughren, J and Pauly, D (2005) A GlobalReview Of Marine Protected Areas. Abstract for the FirstInternational Marine Protected Areas Congress (Geelong,Australia 23–28 October 2005).

Woodroffe, C D, Nicholls, R J, Saito,Y, Chen, Z and Goodbred,S L (2006): ‘Landscape variability and the response of Asianmegadeltas to environmental change’. In Harvey, N (ed)Global Change Implications for Coasts in the Asia-PacificRegion. Springer, Berlin, Heidelberg, New York.

World Bank (ed) (2004) Saving Fish and Fishers. Toward Sus-tainable and Equitable Governance of the Global Fishing Sec-tor. World Bank, Washington, DC.

8 References

Worm, B and Myers, R A (2003) ‘Meta-analysis of cod–shrimpinteractions reveals top-down control in oceanic food webs’.Ecology 84(1): pp162–73.

Worm, B and Myers, R A (2004) ‘Managing fisheries in a chang-ing climate’. Nature 429: pp15.

WPC (World Parks Congress) (2003a) Recommendations ofthe Vth IUCN World Parks Congress. 8–17 September 2003,Durban. (IUCN) The World Conservation Union, Gland.

WPC (World Parks Congress) (2003b) Message of the VthIUCN World Parks Congress to the Convention on BiologicalDiversity. (IUCN) The World Conservation Union, Gland.

WRI (World Resources Institute) (2001) Pilot Analysis ofGlobal Ecosystems. Coastal Ecosystems. WRI, WashingtonDC.

WSSD (World Summit on Sustainable Development) (2002)Plan of Implementation. WSSD, Johannesburg, New York.

Yeung, Y. (2001) ‘Coastal mega-cities in Asia: transformation,sustainability and management’. Ocean & Coastal Manage-ment 44: pp319–33.

Yohe, G, Neumann, J and Marshall, P (1999): ‘The economicdamage induced by sea-level rise in the United States’. InMendelsohn, R and Neumann, J E (eds) The Impact of Cli-mate Change on the United States Economy. Cambridge Uni-versity Press, Cambridge, New York: pp178–208.

Zachos, J, Pagani, M, Sloan, L, Thomas, E and Billups, K (2001)‘Trends, rhythms, and aberrations in global climate 65 ma topresent’. Science 292: pp686–93.

Zhang, K, Douglas, B C and Leatherman, S P (2004) ‘Globalwarming and coastal erosion’. Climatic Change 64: pp41–58.

Zickfeld, K, Levermann, A, Keith, D W, Kuhlbrodt, T, Morgan,M G and Rahmstorf, S (submitted) ‘Experts’ view on risk offuture ocean circulation changes’. Climatic Change.

Zondervan, I, Zeebe, R E, Rost, B and Riebesell, U (2001)‘Decreasing marine biogenic calcification: a negative feed-back on rising atmospheric CO2’. Global BiogeochemicalCycles 15(2): pp507–16.

Zwally, H J,Abdalati,W, Herring,T, Larson, K, Saba, J and Stef-fen, K (2002) ‘Surface melt-induced acceleration of Green-land ice-sheet flow’. Science 297: pp218–22.

Zwally, H J G, Mario, B, Jun, L, Cornejo, H G, Beckley, M A,Brenner,A C, Saba, J L and Donghui,Y. (2005) ‘Mass changesof the Greenland and Antarctic ice sheets and shelves andcontributions to sea-level rise: 1992–2002’. Journal of Glaciol-ogy 51(175): pp509–27.

110

Publications of the German Advisory Council on Global Change (WBGU)

World in Transition: Fighting Poverty through Environmental Policy. 2004 Report.London: Earthscan © 2005, ISBN 1-85383-883-7.

World in Transition: Towards Sustainable Energy Systems. 2003 Report.London: Earthscan © 2004, ISBN1-85383-882-9.

Renewable Energies for Sustainable Development: Impulses for renewables 2004. Policy Paper 3.Berlin: WBGU © 2004, 24 pages, ISBN 3-936191-06-9.

Climate Protection Strategies for the 21st Century: Kyoto and beyond. Special Report 2003.Berlin: WBGU © 2003, ISBN 3-936191-04-2.

Charging the Use of Global Commons. Special Report 2002.Berlin: WBGU © 2002, ISBN 3-9807589-8-2.

Charges on the Use of the Global Commons. WBGU Policy Paper 2.Berlin: WBGU © 2002, 16 pages, ISBN 3-936191-00-X.

World in Transition: New Structures for Global Environmental Policy. 2000 Report.London: Earthscan © 2001, ISBN 1-85383-852-7.

World in Transition: Conservation and Sustainable Use of the Biosphere. 1999 Report.London: Earthscan © 2001, ISBN 1-85383-802-0.

The Johannesburg Opportunity: Key Elements of a Negotiation Strategy. WBGU Policy Paper 1. Berlin: WBGU © 2001, 20 pages, ISBN 3-9807589-6-6.

World in Transition: Strategies for Managing Global Environmental Risks. 1998 Report.Berlin: Springer © 2000, ISBN 3-540-66743-1.

World in Transition: Environment and Ethics. 1999 Special Report.Website: http://www.wbgu.de/WBGU/wbgu_sn1999_engl.html.

World in Transition: Ways Towards Sustainable Management of Freshwater Resources. Report 1997. Berlin: Springer © 1998, ISBN 3-540-64351-6.

The Accounting of Biological Sinks and Sources Under the Kyoto Protocol - A Step Forwards or Backwards for Global Environmental Protection? Special Report 1998. Bremerhaven: WBGU © 1998.ISBN 3-9806309-1-9.

World in Transition: The Research Challenge. 1996 Report. Berlin: Springer © 1997. ISBN 3-540-61832-5.

World in Transition: Ways Towards Global Environmental Solutions. 1995 Report.Berlin: Springer © 1996. ISBN 3-540-61016-2.

World in Transition: The Threat to Soils. 1994 Report. Bonn: Economica © 1995. ISBN 3-87081-055-6.

World in Transition: Basic Structure of Global People-Environment Interactions. 1993 Report.Bonn: Economica © 1994. ISBN 3-87081-154-4.

All WBGU Reports can be downloaded through the Internet from the website http://www.wbgu.de

ISBN 3-936191-14-X

Special Report

R. SchubertH.-J. SchellnhuberN. BuchmannA. EpineyR. GrießhammerM. KulessaD. MessnerS. RahmstorfJ. Schmid

German Advisory Council on Global Change(WBGU)

WBGU



C

M

Y

CM

MY

CY

CMY

K

wbgu_sn2006_deckel_en_2.ai 26.07.2006 13:08:13 Uhrwbgu_sn2006_deckel_en_2.ai 26.07.2006 13:08:13 Uhr

The Future Oceanscgmdesign.fatcow.com/EPBRS/Documents/Reports/Future...mark) (2005): Assessment of...

Documents

Transcript of The Future Oceanscgmdesign.fatcow.com/EPBRS/Documents/Reports/Future...mark) (2005): Assessment of...