Institut du Developpement Durable et des Relations Internationales (IDDRI)
Raphaël Billé , Elisabeth Druel , Julien Rochette ,
IDDRI, Institute for Sustainable Development and International Relations
27, rue Saint Guillaume, 75337 Paris Cedex 07 ? France, www.iddri.org
The 1992 Earth Summit in Rio de Janeiro gave birth most noticeably to the three ?Rio Conventions? (on desertification, climate change and biodiversity) as well as to Agenda 21. Twenty years later, the ?Rio+20? conference has much lower ambitions: in a best-case scenario, it will end up with a strong political declaration, visionary enough to blow a new wind of change and specific enough to open concrete avenues for progress. Officially , ?the objective of the Conference is to secure renewed political commitment for sustainable development, assess the progress to date and the remaining gaps in the implementation of the outcomes of the major summits on sustainable development, and address new and emerging challenges. The Conference will focus on two themes: (a) a green economy in the context of sustainable development and poverty eradication; and (b) the institutional framework for sustainable development?.
This paper aims at contributing to three of these four objectives with regard to oceans by (i) demanding renewed political commitment for sustainable ocean management; (ii) proposing options to fill the remaining gaps in ocean governance; and (iii) addressing some new and emerging challenges. We do not elaborate on the assessment of progress to date, which is clear from UNEP?s Global Environmental Outlooks: in a nutshell the state of the oceans has continuously and seriously worsened since 1992, despite some local or sectoral successes which remain very scattered. Degradation trends are even often still accelerating.
The choice here is therefore to try and contribute directly to the conference objectives, without however developing our views in the framework of the two separate conference themes. The reason is two-fold:
- Oceans have had their share of empty concepts, buzzwords and so-called tools that are actually ?Lichtenberg knives ?. We strongly oppose the supposed need to reframe the problem under a new umbrella, like the ?blue-green economy?. The ocean community is now strong enough, its science robust enough, its concepts and principles established enough, not to let new fashions impose more chewing-gum concepts which bring neither tools, communication, political momentum nor helpful reframing.
- The institutional framework is of course important, but we believe that environmental governance in general, and ocean governance in particular, is bound to remain complex . Efforts at rationalizing the institutional and organizational framework may achieve minor improvements ? and let?s take these! ? but the system will remain essentially multi-layer, heterogeneous, contradictory and seemingly irrational. That is because it was never conceived as a ?system?. Its complexity is the context in which we shall work and make progress, not a temporary disfunctioning which should be remediated.
We argue that significant progress has been made on science ? not a limiting factor at this stage, on the diagnosis of threats and governance gaps, and on legal and policy proposals. It is time to be specific, and the Rio+20 outcome, although a mere political declaration, can help a lot provided it addresses at least five key issues which can perfectly be dealt with outside the green economy ?conceptual? framework. Without any pretention to be exhaustive, we call for progress on (1) setting the stage for the establishment and management of MPAs in Areas Beyond National Jurisdiction; (2) paving the way for an international regulation of offshore oil exploitation; (3) giving the regional approach a new lease on life; (4) giving a new impetus to a drastic reduction of land-based pollution; and (5) strengthening the tools to fight Illegal, Unreported and Unregulated fishing.
2. Setting the stage for the establishment and management of MPAs in ABNJ
Areas beyond national jurisdiction (ABNJ) cover around half of the planet?s surface and 64% of the surface of oceans and seas. They are also the least known and least protected areas on Earth. In recent years, the exponential use of ABNJ and their resources and a growing human pressure have subjected them to a multiplicity of threats: overexploitation of fish stocks, alteration of deep water habitats due to destructive fishing practices, oil pollution, introduction of alien invasive species, acidification of the oceans as well as emerging threats linked to bio-prospecting and deep-sea mining?
In this context, marine protected areas (MPAs) are seen as an efficient tool to protect this fragile biodiversity, but a certain number of legal issues are raised when it comes to the question of their establishment and management in ABNJ .
The United Nations Convention on the Law of the Sea (UNCLOS) is the overarching framework for the governance of the oceans. Its Part VII deals with the legal regime of the high seas and defines its freedoms (freedom of navigation, fishing, marine scientific research?) but seems rather incomplete on the protection of marine biodiversity in this zone. In particular, nothing is said about the establishment of MPAs.
For that reason, regional seas conventions have recently taken the lead on this issue, and created the first MPAs in high seas . But this regional approach has intrinsic limits. First, MPAs designated by a regional sea convention only bind contracting parties to that convention, and not third States, which can nevertheless be active in the region. Moreover, regional seas conventions do not have the mandate to regulate some activities which occur in the ABNJ, such as fishing, deep-sea mining or navigation. Other organisations are competent for these issues, and there is therefore a need of cooperation and collaboration between all these authorities which yields high transaction costs. Last, most parts of the high seas are actually not regulated by a regional convention.
At the same time, the Convention on Biological Diversity (CBD) has adopted scientific criteria for identifying ecologically or biologically significant marine areas in need of protection in open ocean waters and deep-sea habitats (EBSAs). The identification of EBSAs is ongoing, and will probably serve as a scientific basis for the designation of MPAs should the legal framework permit. In 2010, contracting parties to the CBD also agreed in Nagoya to establish a network of MPAs covering 10% of the oceans, including ABNJ, by 2020. October 31st, 2011 Advancing the Oceans agenda at Rio+20: where we must go Raphaël Billé , Elisabeth Druel , Julien Rochette , These three spheres of action (regional, UNCLOS and CBD) are interrelated and need to be strengthened in order to establish an efficient governance system for MPAs in ABNJ. At the regional level, agreements involving inter alia regional seas conventions and regional fisheries management organisations need to be concluded, and complementary models must also be established for parts of the world where no such agreements may exist. Given the growing importance of the CBD due to its prominent position in the identification of EBSAs, its role must also be enhanced. Additional protection could be granted to EBSAs through, for example, a resolution of the United Nations General Assembly (UNGA). However, there is growing evidence and consensus that all these efforts will fall short of delivering adequate protection if a global legal framework does not provide for the establishment and management of MPAs in ABNJ. A clear political mandate must therefore be given at Rio + 20 to the UNGA to open the negotiations on an implementing agreement to UNCLOS on the conservation of marine biodiversity in ABNJ, including through the creation of MPAs. This needs to include other issues such as the status of marine genetic resources in ABNJ or the need of environmental impact assessments, as a package deal. 3. Paving the way for an international regulation of offshore oil exploitation A recent series of accidents on offshore oil platforms (Montara platform, August 2009; Deepwater Horizon, April 2010; Penglai 19‐3 platform, June 2011?) have raised public awareness on the extent to which offshore oil exploitation is moving into increasingly deep waters. Whereas just after the Second World War industries were only drilling in around 10 metres of water, it is now increasingly common for rigs to drill at a depth of over 2 km. Almost a third of the oil consumed in the world now comes from underwater areas. However, human domination of the world?s oceans does not look set to abate. The sea has so far revealed only a tiny fraction of its energy potential and new ultra‐deepwater drilling technologies are currently being developed. Consequently, it is assured that despite its environmental, economic and social impact, these recent accidents are unlikely to halt the rush towards offshore drilling, especially given that the technical cost of deepwater oil drilling has been significantly reduced in recent years and that offshore oil fields are continually discovered. Thus, despite several official declarations advocating for the adoption of a moratorium on deepwater drilling, the economic weight of this sector and the fact that it is currently vital to the equilibrium of the oil market, nobody is seriously considering halting its development. What does this mean in practice? Such strong standpoints against deepwater drilling remain useful and should still be supported and strengthened globally. However, it should be done in parallel with attempts to rethink the regulation of the activity itself. Because of the multiplication of accidents, their dramatic consequences both for marine environment and coastal populations and the ?four Ds, which, from now on, are destined to characterize the offshore oil industry in its search for black gold: Deep, Distant, Dangerous and Difficult 10 ?, it is no more possible to plead for a legal status quo. Agenda 21 argued in 1992 that ?the nature and extent of environmental impacts from offshore oil exploration and production activities generally account[ed] for a very small proportion of marine pollution ?. The Rio +20 process must invite the international community to question the suitability of the current framework for the regulation of offshore oil drilling. Indeed, although UNCLOS imposes a general obligation to protect the marine environment, no international convention specifically sets international standards determining the conditions under which States should issue drilling permits. Moreover, in terms of liability and compensation, oil platforms, floating or otherwise are not covered by international agreements such as the Convention on civil liability for oil pollution damage (1992) or the Convention on the establishment of an international fund for compensation for oil pollution damage (1992). Last, the current governance of offshore oil drilling is in fact largely characterised by self-regulation by operators. It is therefore clear that there are two main gaps in global international law: the first is located upstream ? the absence of an international framework on the conditions under which oil exploration/exploitation is authorised and monitored ? while the second is downstream ? the absence of a global instrument relating to responsibilities and the distribution of damage liability. At the regional level, there is only one real legal instrument aimed at specifically monitoring offshore drilling: the Madrid Protocol for the protection of the Mediterranean sea against pollution resulting from exploration and exploitation of the continental shelf and the seabed and its subsoil, adopted in 1994 within the framework of the Mediterranean Action Plan and entered into force in March 2011. In the North-East Atlantic, the OSPAR Commission has adopted a number of regulations on the matter, but in a more fragmented manner. While hoping that these regulations will be applied by all States and that a common regime will be implemented in these two regions where offshore drilling is rapidly developing, international regulation should not be developed only at the regional level. Indeed, in many regions impacted by the oil industry, such as West Africa, there are major governance problems which impede States to ensure an effective control on offshore platforms and to deal with pollution accidents. Such circumstances can only plead for a comprehensive approach that aims to (i) establish a common set of obligations for States (and operators) to cover the entire process of approval, monitoring, intervention, sanctions and liability regime; (ii) and counterbalance the power of oil companies and their professional organizations with an international legal framework that creates obligations, including reporting, and allows the creation of an international convention secretariat. The current initiatives led by Russia within the G20 framework and Indonesia at IMO push in the right direction. Nevertheless, the Rio +20 process constitutes the opportunity to address this issue in a global and non-fragmented framework: the whole international community must now recognize that Agenda 21 is outdated on this particular topic and therefore pave the way for an international regulation of off shore oil exploitation. 4. Giving the regional approach a new lease on life The 1972 United Nations Conference on the Human Environment led to the creation of the United Nations Environment Programme (UNEP) ?to serve as a focal point for environmental action and coordination within the United Nations system?. In its first session, UNEP made the oceans a priority action area and the UNEP Regional Seas Programme was then initiated in 1974 ?as an action‐oriented programme having concern not only for the consequences but also for the causes of environmental degradation and encompassing a comprehensive approach to combating environmental problems through the management of marine and coastal areas?. Since it was launched, the Regional Seas Programme has proven attractive, as evidenced by the more than 140 States participating in regional seas framework. Because ?not every international environmental problem needs to be dealt with on a global level?, such regional arrangements have often given States the opportunity to go ?closer, further and faster ? than at the global level. However, regional frameworks must now overcome some challenges to retain their value and gain effectiveness within the international oceans governance architecture. First, new frontiers must be conquered. For a long time indeed, regional seas frameworks have remained within the confines of territorial waters and exclusive economic zones. Recently, regional cooperation has extended to the high seas (Mediterranean Sea, North East Atlantic) and coastal zones (Mediterranean Sea, Western Indian Ocean), in a still limited way however. Under the impetus of UNEP Regional Seas Programme, regional seas frameworks must therefore study the political and legal feasibility as well as the scientific and geographic relevance of such extension of the regional cooperation. Indeed, the regional approach can compensate some of the current shortcomings of the global level with regard to conservation of biodiversity in ABNJ. It can also help homogenizing ? to a high common denominator ? the way coastal areas are managed at the national level. Besides, since Rio, many initiatives have been conducted within regional frameworks to update and develop legal instruments of cooperation. Nevertheless, one cannot deny that too many conventions and protocols have been and remain dormant or weakly implemented. It is therefore time to reactivate these instruments, by providing regional frameworks the means of their ambition. More efforts, in terms of funding in particular, must therefore be dedicated to strengthening regional systems, so that they be able to keep the cooperation alive and provide States with technical assistance and support for the implementation of the regional instruments. In this regard, the creation of Regional Activity Centres, such as the ones existing in the Mediterranean and Caribbean Seas, must be promoted. It should be underlined that financial constraints to do so may actually often be more easily removed than usually assumed. For instance, reallocating into trust funds an even small proportion of development cooperation funding that is today dedicated to marine biodiversity conservation through projects, would basically allow to sustain at least one RAC in each regional sea. Special attention should also be given to the cooperation between the regional seas conventions and other sectoral organisations interested in marine conservation. The North‐East Atlantic looks like the most advanced region in this respect. Indeed, the OSPAR commission is currently working on a draft collective arrangement between competent authorities on the management of the six MPAs in ABNJ established in 2010. This process, also referred to as the ?Madeira process?, aims to coordinate the adoption of sectoral management measures and to strengthen cooperation on this issue with organisations such as the North East Atlantic Fisheries Commission, the IMO, the International Seabed Authority (ISA) or the International Commission for the Conservation of Atlantic Tunas (ICCAT). Such cooperation between regional and international organisations is also crucial in parts of the world where no regional seas convention and/or no RFMO (regional fisheries management organisation) exist. In the Sargasso Sea for instance, there is no regional sea convention and the only competent RFMO, ICCAT, is only competent for tuna. Regional cooperation can nevertheless be enhanced by a combination of protective measures adopted by sectoral organisations, such as the designation of a particularly sensitive sea area by IMO, of fisheries closures by ICCAT, of areas closed to seabed mining by ISA. In the future, an agreed management plan could be signed by these organisations, and could also enlist the support of coastal States to this initiative. The Sargasso Sea Alliance is currently promoting this approach. These two models show the need of a better cooperation at the regional level but also highlight that there are already instruments existing for this cooperation to be enhanced. At the regional level, States have the power to act without necessarily going through an important institutional reform. Rio +20 is therefore the appropriate moment, not only to recall the strategic importance of the regional approach in addressing marine issues, but to help regional organisations enter the 21st century: the Rio+20 outcomes must urge the international community, including donors and NGOs, to strengthen the regional frameworks so that they be able to make a real difference in tackling marine challenges. As far as high seas conservation is concerned, a mandate given to the UNGA to start a negotiation process towards an UNCLOS implementing agreement should necessarily assert the importance of the regional level, and encourage negotiating parties to strengthen the coordination and cooperation between regional organisations. 5. Giving a new impetus to a drastic reduction of land-based pollution Land-based activities have been recognised as a major source of marine environmental pollution for decades. During the 1992 Earth Summit, States considered that these activities contributed to ?70 per cent of marine pollution?. Even if the Johannesburg Plan of implementation called States to ?make every effort to achieve substantial progress (?) to protect the marine environment from land‐based activities?, the situation has further deteriorated. It is now estimated that up to 80 per cent of marine pollution comes from land‐based sources, which also poses huge threats to human health. Despite the various legal instruments and policies adopted at global, regional and national levels, the situation is therefore getting worse as illustrated by the two related ?emerging issues? identified by the 2011 Report of the UN Secretary‐General on oceans and law of the sea. The first issue relates to the discovering of high levels of accumulation of plastics and other marine debris in seas convergence zones, known as ?gyres?. In the North Pacific for example, studies show a ?literal blanket of trash? while recent expeditions highlighted the presence of 250 billion particles of micro-plastic in the Mediterranean. The ocean is therefore sick from land-based activities, as also illustrates the second issue identified by the UN Secretary General?s report, namely nutrient over‐enrichment of marine waters, which causes oxygen depletion, eutrophication and lead to the development of ?dead zones?. In this regard, dead zones in coastal areas have doubled in extent every decade since the 1960's. A recent study conducted by UNEP has identified 415 eutrophic and hypoxic coastal systems, including 169 identified hypoxic areas, 233 areas of concern and 13 systems in recovery. These two phenomena can obviously be seen as ?emerging issues? but they are, above all, the symptoms of a continuously deteriorating situation. However, policy and legal responses to address land-based pollutions have been widely adopted in recent years, through the elaboration of the Global Programme of Action for the Protection of the Marine Environment from Land-based Activities, the adoption of the European Union Water Framework Directive and the EU Marine Strategy Framework Directive, and specific protocols within regional seas frameworks, as well as the related development of national policies and laws. But one cannot fail to notice that these political and legal instruments have not curbed the general tend. Rio+20 process and outcomes must therefore not miss this issue but recall the extent of the problem. In particular, considerable progress could be made in the future on at least three dimensions. First, there is unquestionably a need to ensure the effective implementation of legal instruments, especially at the regional level, which presupposes the strengthening of the regional systems themselves (see section 4). Second, investments should be reinforced to develop sewage treatment plants, which are dramatically lacking in many coastal zones, and the associated water management systems; donors should thus reinforce this area of action. Last, economic instruments should be used to promote environmentally friendly behaviours and activities: taxes, non‐compliance fees and, more than anything and following the Nagoya Strategic Plan, elimination of harmful subsidies which are supporting models of agriculture, industry, mining or tourism that have been polluting our oceans for too many decades. No major step forward can be expected from Rio+20 in that regard. However simply placing the fight against land-based pollution back among the highest priorities for action and funding would be useful for future reference. 6. Strengthening the tools to fight Illegal, Unreported and Unregulated (IUU) fishing Chapter 17 of the Agenda 21 promoted the sustainable use and conservation of marine living resources and described a wide range of tools available to States in order to achieve this aim. In 2002, at the WSSD, States agreed to ?maintain or restore stocks to levels that can produce the maximum sustainable yield with the aim of achieving these goals for depleted stocks on an urgent basis and where possible not later than 2015?. Today, in 2011, it seems clear that these objectives, formulated with wisdom in very vague terms (?and where possible not later than 2015?), will not be met ? whatever this should mean. The 2010 report of the FAO on the state of the world?s fisheries and aquaculture indeed estimates that about 32 percent of world fish stocks are overexploited, depleted or recovering and that 53 percent of the remaining stocks are fully exploited. The report moreover states that ?the increasing trend in the percentage of overexploited, depleted and recovering stocks and the decreasing trend in underexploited and moderately exploited stocks give cause for concern?. At the same time, market demand for fish has never been so high and is not likely to decrease in the upcoming years. Illegal, Unreported and Unregulated (IUU) fishing is a widespread phenomenon which largely contributes to the depletion of fish stocks. A report to be adopted soon by the European Parliament estimates that its share is equivalent to at least 15% of worldwide catches, between 11 and 26 million tonnes a year. In 2001, the international community decided to tackle the issue with the adoption of an International Plan of Action to prevent, deter and eliminate IUU fishing (the FAO IPOA IUU), an instrument which provides States with a range of tools and principles which could help to fight IUU fishing. The following year, the Johannesburg Plan of Implementation (JPOI) called upon States to ?urgently develop and implement national and where appropriate, regional plans of action to put into effect [?] the International Plan of Action to prevent, deter and eliminate illegal, unreported and unregulated fishing by 2004?. But, since then, only very little progress has been made. The biggest flaw of the FAO IPOA IUU is that it is a voluntary instrument and to date only a small number of States have adopted their own national plan of action. This proves that it might not be the appropriate tool. On the whole, efforts are far from sufficient and, as of today, it is clear that the fight against IUU fishing is a triple failure: - It is the failure of international organisations (RFMOs but also the UN, with UNCLOS and the other agreements linked to it as well as the FAO) to impose an effective control over States on this matter. RFMOs have established IUU lists of fishing vessels, which are all out of date, empty or never contain the names of vessels flagged in one of their Contracting Parties. A reform of the RFMOs is urgently needed in order to strengthen their capacity to ensure compliance with the measures they are adopting and to impose sanctions on States which are not respecting them. - It is the failure of States to exercise an effective control over their own vessels and/or nationals because of a lack of means, but also and primarily because of a lack of political will. The lack of political will is here not only a vague reluctance but in some cases a clear and active political choice, with the existence of ?States of non-compliance?. The international community should adopt means such as a flag State compliance instrument with stringent sanctions associated in case of non‐respect to make sure that these States can no longer escape their international obligations. In addition, and according to the requirements of the JPOI and Nagoya plan of action, States should also eliminate subsidies which contribute to IUU fishing. - It is the failure of the markets to regulate fish exchanges so as not to accept in the production chain products stemming from illegal catches. IUU fishing is a very profitable business and it is relatively easy for operators to take advantage of the system. Trade market related measures have been advocated for a long time, and the EU is pushing for the introduction of a global catch certification scheme. In order to be efficient and to avoid fraud, this scheme should not be paper based, but fully electronic, based for example on the TRACES system used by the EU for health and sanitary certification requirements. More broadly, a particular emphasis must be put on the fight against flags of convenience and in particular against countries which are granting their flags and international fishing licenses (the owner of the vessel is not allowed to fish in the EEZ of the flag State but only in the high seas) to beneficial owners who are hiding their identity under the status of an International Business Company or Corporation, as is currently the case in Belize for example. Parallel registration of fishing vessels should also be carefully monitored. To this end, the establishment of a mandatory record of fishing vessels under the auspices of the FAO would be an efficient instrument. Additional efforts have also to be put on the definition of the link between a flag State and its vessels. Article 92 of UNCLOS simply states that ?there must exist a genuine link between the State and the ship?. An agreed international definition of this link must be adopted, which should underline the necessity of States to effectively exercise their jurisdiction on their vessels. Rio+20 should be the occasion for States to renew their commitment to fight IUU fishing but also to take into account the shortcomings of this fight in the last 20 years. A call is needed for the adoption of stringent international instruments (a global record of fishing vessels, a flag state compliance instrument) and for the ratification of the FAO Port State Agreement. Discussions must be launched on the review of the RFMOs and on the adoption of an international, electronic-based, catch certification scheme. 7. Conclusion We reviewed five key areas where we think steps forward could and should be taken at Rio+20. Readers may be stricken by two observations: 1. These all sound like old stories: none of these issues ? except to a certain extent offshore energy ? is new, and most of the policy responses are well-known. 2. The range of progress that may be achieved in Rio in June 2012 seems ridiculously bureaucratic: launching processes, reasserting the importance of ?, etc. We believe both are realistic and legitimate concerns. First (1), most issues have indeed been around for at least 20 years, often many more, and the international community ? or part of it ? has been trying to take action for almost as long. Rio+20 is also Stockholm+40, and a lot was already going on long before 1972, both in terms of problems and action. Pretending things have been moving in the right direction, although too slowly, would not only be misleading: it would be a strategic dead-end. The only way forward is to recognize the overall failure with regard to ocean governance, to study the few successes at hand, and to develop strategies that seriously take both into account. This means also acknowledging the conflictual dimension of ocean governance and the widespread reluctance ? if not active resistance ? to make it more sustainable as soon as it comes at a cost. The green economy concept, however blue we may manage to make it, does not contain any silver-bullet to overcome such oppositions. We are left with good old advocacy, communication, education, demonstration, to hopefully build the necessary constituency ? hence the political will ? that will reverse the balance of powers. Besides intensification of well-known threats, what may be considered new ? in the sense that we are only starting to understand it ? is the additional pressure climate change and ocean acidification place on marine ecosystems. Impact science is in its infancy, and effects of e.g. acidification are difficult to isolate from that of other stressors. There are however three points we can make: - Climate change and ocean acidification may well turn out to be the greatest threats to marine ecosystems along the 21st century, but this will largely depend on the way we manage other stressors. -As global issues (i.e. issues which need to be addressed globally), they will however impact marine ecosystems and dependent societies very unevenly and at different time scales. This will challenge the willingness to and attempts at addressing these threats globally. - Most importantly, oceans are faced for the first time with global threats rising after most levers and policy options to tackle them have already been identified and tested: reducing CO2 emissions, building ecosystem resilience through networks of MPAs, the reduction of land-based pollutions or reducing the pressure from fishing, etc. Although some minor progress has been made along these lines, efforts have remained vastly insufficient and unsuccessful. This means that there is not much choice but to eventually succeed where we failed to a large extent so far. Hence the old stories and familiar policy responses reviewed in this paper. Second (2), having too high expectations about Rio+20 would not help. The above indeed shows the nature of what may be achieved there: at best the international community may decide to renew and strengthen the political momentum inherited from Rio 1992, and to launch new processes which in turn may lead to important decisions. But crucial decisions ? in the sense of decisions directly driving changes in pressures on oceans ? will not be taken in Rio. It is all the more important to demand that Rio+20 be a political and bureaucratic success. This will condition to a significant extent the next decade of the international sustainable development agenda. Powerful International Science ? Policy Interfaces for Sustainable Development: organise their proliferation, accept and clarify their political role Sébastien Treyer, Raphaël Billé , Lucien Chabason, Alexandre Magnan IDDRI, Institute for Sustainable Development and International Relations 27, rue Saint Guillaume, 75337 Paris Cedex 07 ? France, www.iddri.org This contribution for UNCSD 2012 concerning the Institutional Framework for Sustainable Development builds upon IDDRI?s experience in supporting negotiations about the establishment of international science‐policy interfaces like IPBES , and on recent publications and workshops intending to take stock of the variety of science-policy interfaces having emerged for the international governance of the environment or other sustainable development objectives like food security. 1. Organising the proliferation International Science-Policy Interfaces (ISPIs) are proliferating in the various regimes of international governance of sustainable development (scientific advisory bodies of various environmental conventions, IPCC , IPBES, World Water Assessment Programme, High level panel of experts for Food security and nutrition, Assessment of Assessments [AoA] for the marine environment, foreseen panels on soils at FAO or for the UNCCD ?). Their multiplication is for the moment often based on the replication of successful mechanisms or experiences, that are used as references : the IPCC has obviously been one of them, but also the Millenium Ecosystem Assessment as far as the IPBES is concerned; other main references are the role played by science in addressing long range transboundary air pollution (LRTAP / Acid rains), tropospheric ozone depletion (Montreal protocol), or environmental issues in the Mediterranean (Barcelona Convention). Integrated modeling exercises and integrated assessment methods (scenario and simulation, for instance) have been co-evolving with these different exercises, playing a central role in the mechanisms that are acknowledged as successful. Influential personalities and key research institutions have also played a central role in transferring experiences from one field to another. There is therefore some genealogy underpinning the multiplication of ISPIs . But recent experiences and publications put the stress on the specific political context and specific structure of academic and epistemic communities in each field, showing that the role of science has to be analysed specifically in each case. Recommendations 1a. It is useful to seek for synergies and coordination, in order to avoid duplication among ISPIs. 1b. The role of science in each field has to be identified specifically, and an existing ISPI should not be directly substituted to the expected function in another field, neither can the model of an existing ISPI be directly replicated in another field. 1c. Having identified a specific function for scientific expertise in an international policy process should not necessarily lead to the institutionalisation and the design of a new ISPI. Empowering existing academic arenas and coordinating partial, scattered existing mechanisms should also be considered valid options. 1d. Drawing lessons from experiences of ISPIs in other fields is useful and should be organised, without leading to direct replication. 2. Accepting and clarifying the power of ISPIs comes first, before ensuring their efficiency Current discussions about ISPIs are overwhelmed by questions concerning the improvement of their efficiency, and the corresponding optimal institutional design. As stated before, questions of institutional design should only come after the identification of the function that is expected from scientific expertise in the specific area under consideration. The reference models of successful ISPIs that are generally used to define efficiency improvements have to be put into perspective, as they often refer to situations where the key role played by science was to influence the policy agenda by making the case for a specific environmental problem, for which technological solutions were apparently already at hand, which was the case for LRTAP or the Montreal protocol. It is in the same perspective that the IPCC, and in particular its Working Group I on climate science, have been seen as a reference, because of their very efficient process for clarifying the characteristics of climate change, giving evidence of its anthropogenic cause, and making a clear statement of the minimum expected level of climate change during the next century. In such a perspective, it is generally considered as the ideal model that scientists should reach consensus and clarify residual uncertainties in order to have an impact on decision makers. Efficiency is then a question of: ‐ ensuring good quality standards for the production of ?sound science?; ‐ improving communication channels and mechanisms to reach policy makers. But this perspective is only relevant: ‐ for experimental sciences and biophysical phenomena; ‐ and if the objective is to put an environmental issue on the agenda. For many other scientific fields and policy contexts, controversies are inevitable (whenever it comes to social changes, development pathways, and to social sciences or economics), and they can not and should not be reduced to consensus. And for many decisions, the role of science is not anymore in agenda setting, but rather on comparing policy options. Working group III of IPCC provides an interesting example of producing scientific expertise on possible socio-economic trajectories to 2100 in order to inform a negotiation process where burden sharing has been at the heart of the discussions. Solutions found to deal with divergences among researchers in WG III are nevertheless much less analysed and discussed than WG I processes, although they might represent very useful experiences. Food security is also an interesting domain, for which the critical function of science might well be much more in the evaluation of the performance of public policies and regulations than in putting food security on the international agenda. The role of science for the evaluation of policies will be all the more critical when the focus comes down from international regulations (what options do we have to regulate the volatility of international prices ?) to assessing the performance of national domestic policies, that are essential drivers of food security. Such an evaluation however raises sovereignty issues and will therefore necessarily question the legitimacy of the ISPI to work on this core question. Recommendations 2a. The political function of science should be acknowledged, particularly as it should not be reduced to producing biophysical scientific evidences for the only purpose of agenda setting. 2b. The role of science, and the function of ISPIs if the institutionalisation is seen as necessary, has to be considered as strategic, in so far as knowledge production ensured in these arenas has a key role to play in advocating for changes in current policies, by, inter alia: ‐ assessing the current state of a problem; ‐ evaluating the performance of existing or possible policy options; ‐ evaluating the costs (not only economic costs, but also social, environmental, political costs) of the different options, and the distribution of these costs among the different negotiating parties or non negotiating stakeholders; ‐ closing controversies when possible; ‐ re‐opening the range of possible options; ‐ re‐framing the formulation of a problem to include dimensions that are not enough taken into account? This first list of possible functions of scientific expertise in international negotiation processes should illustrate that there are a diversity of possible strategic roles of science. They have to be identified carefully, in order to clarify how to ensure the legitimacy of science, or the corresponding ISPI if considered necessary, for such a strategic role. 2c. The role of science is therefore of a political nature, but it will never be to impose any decision calculated as optimal, which would substitute expertise to the negotiation process. Science should not be expected to take over the responsibility of political decision‐makers and negotiators. Improving the legitimacy of ISPIs is important, and can necessitate pluralism, or even the participation of a variety of stakeholders because they hold a key component of relevant knowledge, as is the case in IPBES. However such a ?democratisation? process in the field of knowledge production should never result in its substitution to political decision making. 2d. Common criteria used to assess the efficiency of ISPIs (credibility, legitimacy, relevance) are very useful tools to improve the capacity of science or of the corresponding ISPI, but only once the strategic role of science has been clarified. Depending on the context and the expected function, credibility should not directly be reduced only to quality processes to ensure ?sound science?. It can be linked to the organisation of a pluralistic academic debate among different worldviews (for instance, in economics, about the impacts of trade liberalisation on socio‐economic development), showing both convergences and divergences in the existing scientific production, that the decision maker has to bear in mind when making a political choice. It would not be science?s role to substitute for the political choice by a false consensus. Independence from any mandatory body, be it national or international, can be a very important criterion to ensure some of the strategic functions science could play (the capacity to re‐open the range of possible options, for instance, necessitates that the mandate of an ISPI is not too narrowly defined by its mandatory body), and it would increase its credibility and also its legitimacy. But in some intergovernmental negotiation processes, legitimacy might need to be ensured through some kind of intergovernmental designation of experts, and relevance might necessitate some definition of the mandate by an intergovernmental process. In every specific context, it is thus necessary first to analyse and specify the strategic function that is expected from science, or from an ISPI if considered necessary, in order to clarify its political role, and to address specifically how to ensure or improve credibility, legitimacy and relevance, with respect to the specific expected strategic function. It is not just efficient but really powerful science-policy interfaces that are needed for the international governance of sustainable development: ensuring they can have power, in a legitimate and efficient way, is more than a question of institutional design. UNCSD Rio+20 should be the opportunity that the international community endorses the necessity to empower each ISPI with a strategic and political role, although this role differs depending on the issue at stake. A process should be launched to enable to specify this strategic role in each case and at the same time draw relevant lessons from other experiences without replicating one-size-fits-all models.