Version classiqueVersion mobile

Climate Technologies as Emergency Solutions

Pascal Blickle

Chapter 2: A Survey of Three Proposed Technological Solutions for Climate Emergencies1

Texte intégral

  • 1 Parts of the discussion of artificial islands in this chapter are based on a paper entitled “We’re (...)

1This chapter is devoted to the study of three technologies for which an emergency framing is often employed. These case studies aim to examine how the legal rules applicable to these technologies address their potential status as emergency solutions to climate change. I will examine the different roles and relationships that relevant legal rules may have with respect to the various technologies. Three technologies have been chosen for this analysis: stratospheric aerosol injection (SAI), ocean fertilisation, and artificial islands. Two of these case studies, stratospheric aerosol injection and ocean fertilisation, are often presented as prime examples of geoengineering (IPCC 2012, 1-7; Shepherd et al. 2009, 9-36). Conversely, artificial islands are not commonly included in the geoengineering literature. However, I will attempt to show that there are common traits between all three technologies, notably through their shared emergency framing. As discussed previously, analysing geoengineering technologies as entirely distinct from other climate technologies – as the literature often does – might not necessarily be suitable for studying emergencies and climate technologies in international law.

2The chapter will first provide a short overview of each of the three studied technologies and outline their similarities and differences. It will then present, in turn, law that is prohibitive towards the technologies, calls for precaution, is ambiguous, encourages the technologies, facilitates funding and research for them, and creates fora for their future discussion and regulation. As not all legal rules potentially applicable to these technologies can be studied, the most relevant ones have been chosen for this study. Instruments whose organs have already explicitly dealt with such technologies or where such engagement is probable in the future are prioritised in particular.

1. Introducing the Case Studies

1.1 Stratospheric Aerosol Injection

3Stratospheric aerosol injection (SAI) is part of a larger set of technologies known as solar radiation management (SRM). SRM technologies aim at creating a cooling effect by diverting incoming solar radiation (Shepherd et al. 2009, 23). SAI attempts to obtain such diversion by the injection into the stratosphere of particles that reflect solar radiation (Crutzen 2006, 212; Shepherd et al. 2009, 29). This injection could be carried out by different means, such as aeroplanes, rockets, artillery, or balloons (Rasch et al. 2008, 4013-4015; Shepherd et al. 2009, 32). To date, most research on SAI has focused on sulphate aerosols as the substance to be injected into the stratosphere. These aerosols are already naturally present in the stratosphere and contribute to the earth’s natural albedo (Rasch et al. 2008, 4009; Shepherd et al. 2009, 29). The literature often refers to volcanic eruptions as an analogue of sulphate-based SAI, as such eruptions propel sulphur particles into the stratosphere and have a proven cooling effect on the planet (Crutzen 2006, 212; Rasch et al. 2008, 4009). However, the actual effects of SAI might differ significantly from volcanic eruptions, as it would involve continuous injection over decades or centuries, most likely in several locations, rather than a one-off, localised eruption (Rasch et al. 2008, 4009; Shepherd et al. 2009, 29).

4SAI has been present in scientific debates for several decades. In 1974, Russian climatologist Mikhail Budyko proposed that stratospheric sulphur injections by aeroplanes could be used as a remedy if climate change became a serious threat (Rasch et al. 2008, 4008; Stilgoe 2018, 680). Nonetheless, the idea was virtually absent from any serious discussion of climate change policy until the climatologist and Nobel prize laureate Paul Crutzen published an article calling for SAI research in 2006 (Crutzen 2006. See Hällström 2008, 376; Stilgoe 2018, 680). In recent years, SAI has received increased consideration in political settings. For example, in 2020, the United States Congress approved funding for research into SAI and other SRM technologies and discussed authorising government agencies to conduct atmospheric experiments (Fialka 2020). Today, SAI, and more specifically the injection of sulphate or sulphur dioxide, is considered the most probable candidate for research and development among SRM technologies and has received the most media attention (Anshelm and Hansson 2014a, 137; Muffett and Felt 2019, 36; Reynolds and Fleurke 2013, 102). Some field research has already been carried out with certain SAI technologies, albeit on a relatively small scale and focused on a limited set of research questions (Doughty 2019). This comparatively high number of real-life experiments is undoubtedly partly due to SAI’s low cost and flexibility of deployment (Lin 2013, 194).

5As SRM technologies neither aim to reduce greenhouse gas concentrations in the atmosphere nor adapt to the effects of climate change, the IPCC considers them not to fall into the categories of mitigation and adaptation (IPCC 2012, 2-3). Due to the significant uncertainties around SRM, the IPCC does not consider these technologies in its potential pathways (IPCC 2018, 12-13). Moreover, there is a range of potential risks associated with the deployment of SAI technologies (Robock 2008). For instance, SAI might severely and disparately impact regional climates, notably by modifying precipitation patterns, impacting human food production (Robock 2008, 15; Shepherd et al. 2009, 31). The ozone layer is also likely to be harmed by the injections of aerosols, which could lead to an increase in ultraviolet radiation reaching the earth and potentially affecting human health (Rasch et al. 2008, 4030; Robock 2008, 15-16; Shepherd et al. 2009, 32). The injection of sulphur might also contribute to acid rain. However, as quantities of injected sulphur would likely be relatively small compared to industrial emissions, the danger might be more one of new regions being affected by acid rain than of an overall increase in the phenomenon (Rasch et al. 2008, 4032; Robock 2008, 16; Shepherd et al. 2009, 31-32). Finally, SAI only addresses some of the consequences of high atmospheric GHG concentrations, notably failing to reduce ocean acidification (Rasch et al. 2008, 4009-4010; Robock 2008, 15).

6Other risks of SAI include political risks linked to governance as well as misuse of the technologies (Robock 2008). Governance risks are mainly linked to the relatively low costs that would make it possible for a single state or a group of states – or even private companies – to launch SAI without consulting other affected states (Bodansky 2013, 547-549). Furthermore, once SAI technologies are deployed, stopping continuous or regular injections would entail a sudden spike in temperatures corresponding to the levels of GHGs that would have increased further during the SAI deployment period (Shepherd et al. 2009, 24). Through this so-called termination problem, the chosen technology would constrain future human agency, as it would render the technology’s cessation an impossible choice (Fleurke 2017, 214).

7Legal instruments from different fields are pertinent to SAI. Climate change law acts as an encouragement for the technology but also calls for precaution and could potentially provide a forum for future discussion and regulation. Instruments on the protection of the atmosphere generally call for precaution, notably the Convention on Long-Range Transboundary Air Pollution (CLRTAP) and the International Law Commission’s (ILC) draft guidelines on the protection of the atmosphere.

1.2 Ocean Fertilisation

8Ocean fertilisation has been described as the most advanced type of geoengineering, with a number of field experiments having already been conducted (Ginzky and Frost 2014, 83; Verlaan 2009, 447). Thanks to the work of the London Convention and London Protocol, it is also the first type of geoengineering regulated by a legally binding instrument, even though this instrument has not yet entered into force (Ginzky and Frost 2014, 82). On a technical level, ocean fertilisation aims at increasing the uptake of atmospheric CO2 by oceanic algae, which then transport the carbon to deeper waters, where the CO2 is sequestered in the long term (Shepherd et al. 2009, 16-17; Williamson et al. 2012, 476). The technology involves adding certain nutrients that increase the growth of algae and other photosynthetic organisms. While adding nitrogen or phosphorus have been proposed, most proposals and experiments focus on the addition of iron in the form of ferrous sulphate (Shepherd et al. 2009, 16-17; Williamson et al. 2012, 476-477).

9Since the idea of ocean iron fertilisation was first proposed in 1990, at least a dozen experiments have been conducted in oceanic regions around the world (Scott 2018, 40; Williamson et al. 2012, 477-478). These experiments and further research have identified a range of potential side effects. The potential unintended consequences include increased production of other GHGs such as methane and nitrous oxide, decreased oxygen levels and increased acidification in lower ocean regions, and increased production of toxic algae (Shepherd et al. 2009, 17-18; Williamson et al. 2012, 480-482). Furthermore, results from experiments question the technology’s efficacy. Ocean fertilisation’s potential to sequester carbon was shown to be significantly lower than expected, meaning that even global-scale efforts could only address less than 10% of global GHG emissions (Williamson et al. 2012, 479). Moreover, it is unknown how much sequestered CO2 will be returned to the atmosphere over the long term through the natural upwelling of deep waters (Williamson et al. 2012, 480).

10Two factors make regulating ocean fertilisation particularly challenging. First, this activity is destined to take place on the high seas. Ocean fertilisation would probably not be effective in coastal waters, as these waters already contain many nutrients from land-based sources, and additional fertilisation would thus not have any significant effect on algal growth and GHG absorption (Branson 2014, 176; Schlacke and Kenzler 2009, 757). The implementation of any regulation on ocean fertilisation therefore has to rely on flag and port state jurisdiction (Rayfuse, Lawrence and Gjerde 2008, 317-320). Second, ocean fertilisation could be conducted to a large degree by private corporations. Several companies have already attempted to commercialise ocean fertilisation, as the technology is readily available and promises the cheap generation of carbon credits (Eick 2010, 360; Freestone and Rayfuse 2008, 227-228).

11Currently, the law of the sea, the law on biological diversity, and climate change law are particularly relevant for ocean fertilisation. The first two areas of law adopt a precautionary stance towards ocean fertilisation and have expressed this through several non-binding and binding instruments that refer explicitly to the technology. By contrast, climate change law is more ambiguous in its treatment of ocean fertilisation, offering both encouraging and cautioning elements.

1.3 Artificial Islands

12The idea of constructing artificial islands in the oceans is not new. Already in the last century, plans for such islands were proposed and partly realised on a small scale for different purposes, including for escaping taxation or regulation of gambling or radio broadcasting (Walker 1972, 638-641). Between the 1950s and 1970s, large-scale projects involving the construction of whole cities or industries on artificial islands were considered by certain states to respond to population growth (Papadakis 1977, 16-17; Wang 2019).

13More recently, artificial islands or “floating cities” have been examined by some states as a potential solution to problems caused by climate change-induced sea-level rise (Oral 2019, 430). Seen as a climate change measure, the construction of artificial islands responds to a different aspect of “climate emergency” than the previous case studies. Rather than addressing greenhouse gas concentrations or temperatures, artificial instruments are proposed as an adaptation measure to sea-level rise (Oral 2019, 417-418). This adaptation measure aims at replacing the disappearing territory of low-lying states to ensure the continued physical and legal existence of the territory, the basis of life they provide, and the state (Oral 2019, 430-431). Accordingly, the emergency that is being responded to is sea-level rise rendering (island) territories uninhabitable.

14Different types of constructions may be considered artificial islands. Notably, both structures built directly on the seafloor with natural materials and installations floating on the sea surface can be subsumed under this term (Grote Stoutenburg 2015, 169-170; Yamamoto and Esteban 2014, 159-167). The Maldives has already begun erecting an artificial island of the former type, with residential and commercial zones next to its main island, which might serve as a “replacement” island in the future (Republic of Maldives 2019, 8. See also Gagain 2012, 86-87; Hinkel et al. 2018, 573-574). Moreover, the government of the Maldives and a Dutch company plan to start constructing a floating city in 2022 (Marchant 2021). Similarly, the president of Kiribati has announced that he is contemplating acquiring floating islands to relocate the population of his country (Grote Stoutenburg 2015, 169). Artificial islands for economic purposes, particularly for tourism, have been erected in different parts of the world, most famously in Dubai (Rayfuse and Crawford 2011, 11). A range of other ambitious but mostly vague plans for building floating cities to replace sinking territories and house climate refugees are advocated for by different groups, some with support from UN agencies (Callebaut 2015; Ranganathan 2019; UN-Habitat 2019).

15While artificial islands might contribute to lessening some of the impacts of climate change on human life, both their construction and their continuous presence in the ocean are likely to negatively impact the marine environment (Portman 2019, 160-161). Artificial islands may notably cause changes in currents, temperatures, salinity, and light, which would impact marine plant and animal life and might also lead to increased erosion elsewhere (Portman 2019, 161; Gagain 2012, 119-120). Moreover, the inhabitants of these new islands are likely to cause additional stress on the marine environment through their economic activities, mainly through the generation of waste and the use of fisheries and marine plant life (Portman 2019, 161). Finally, the relocation of an island’s population onto an artificial structure is likely to have socioeconomic impacts on the population; the availability of affordable housing and the viability of indigenous lifestyles are likely challenges (Bryant-Tokalau 2018, 38).

16The law of the sea, the law on statehood, and climate change law are likely to play a role in determining the future of artificial islands as a project to adapt to climate change. None of these three areas of the law has a clear stance on artificial islands. The law of the sea and the law on statehood do not clearly answer the question of whether such constructions could fulfil their purpose of preserving the legal existence of sinking states. Different parts of the law on climate change similarly send mixed signals about the legal regime’s availability to facilitate such projects as adaptation measures.

1.4 Similarities and Differences

17The three technologies have been selected as case studies because they are all frequently proposed by their proponents as emergency solutions to climate change. They are presented as an inevitable plan B in the face of current insufficient mitigation and adaptation. At the same time, these technologies and their emergency framing have been criticised for distracting from mitigation and adaptation efforts. While the three case studies all share this emergency framing, they distinguish themselves in other regards.

18The three technologies do not occupy the same place in climate change policies. Ocean fertilisation and other technologies aiming at removing greenhouse gases from the atmosphere can be considered mitigation measures – or at least as pursuing a similar purpose to mitigation (Heyward 2013, 24-25; IPCC 2012, 3). Considering ocean fertilisation as a mitigation measure appears to be in line with the IPCC’s definition of mitigation, as it includes the increase of carbon sinks in its scope (IPCC 2014, 76). SAI may be placed between mitigation and adaptation, as it aims at addressing neither the cause of climate change – greenhouse gas emissions and concentrations, addressed by mitigation – nor the consequences of climate change, addressed by adaptation, but acts at an intermediate step, attenuating temperature rise by influencing incoming solar radiation (Heyward 2013, 25; IPCC 2012, 3). Finally, artificial islands as a replacement for territory rendered inhabitable by climate change-induced sea-level rise can most aptly be described as an adaptation measure (Oral 2019). Following the IPCC’s definition of adaptation, building artificial islands as a response to the loss of inhabitable territory can indeed be considered “adjustment to actual or expected climate and its effects in order to either lessen or avoid harm”, as the construction of these structures aims at lessening the harm caused to people losing their territory to sea-level rise (IPCC 2014, 76).

19Moreover, the legal and technological development of the three technologies is not at the same stage. While computer models have been developed for all three technologies, ocean fertilisation is the farthest in the technological development process, with over a dozen field experiments already conducted. On the legal front, regulation is equally most advanced for ocean fertilisation, as states have been active in adopting regulation in the frameworks of the London Convention and Protocol (LC/LP) and the Convention on Biological Diversity (CBD).

2. The Different Roles of International Law

2.1 Prohibiting Law

20Currently, there are no instruments in international law that specifically target one of the technologies discussed here with prohibitive rules. As the three technologies mostly use substances and equipment that are not prohibited or severely regulated, few prohibitive provisions would hinder their deployment. This absence of explicit regulation is not particularly surprising, given that most relevant instruments were negotiated when technologies for the intentional alteration of the climate were not yet imaginable (Bodansky 1996, 316; Virgoe 2009, 109).

21SAI could be subject to the prohibitive provisions of certain protocols to the Convention on Long-Range Transboundary Air Pollution (LRTAP Convention or CLRTAP) if sulphur is used as the injected substance. This regional framework convention and its protocols address air pollution with transboundary effects. It was developed following concerns about the acidification of surface waters and acid rain and was adopted in 1979 by the United Nations Economic Commission for Europe (Dupuy and Viñuales 2018, 150-151). Today, the Convention counts 51 parties, including virtually all Western European states, the US, Canada, and Russia. The Convention’s scope is limited to long-range transboundary air pollution, which is defined as pollution having its effects in another state, while the contributions of individual sources cannot be distinguished due to the long distance (art. 1(b)). It is noteworthy that the pollution necessarily has to affect another state; pollution affecting only the environment outside state jurisdiction is beyond the CLRTAP’s scope (Dupuy and Viñuales 2018, 152).

22The Convention has several protocols that regulate a specific set of substances or specific problems caused by a group of substances (Dupuy and Viñuales 2018, 153-158). Three of these protocols, namely the Helsinki, Oslo, and Gothenburg Protocols, address sulphur emissions. Their geographical scope is limited, as only some of the (already geographically constrained) CLRTAP parties have acceded to the protocols (Brent 2017, 42). While the Helsinki and Oslo Protocols exclusively address sulphur, the Gothenburg Protocol addresses various substances that may cause acidification and similar problems (Helsinki Protocol, art. 2; Oslo Protocol, art. 2(1); Gothenburg Protocol, art. 2(1)). The Helsinki Protocol provides a single reduction goal for all states to be achieved in a set timeframe (art. 2). Contrarily, the Oslo and Gothenburg Protocols provide that emissions should not exceed “critical loads”, which are defined as quantities “below which significant harmful effects on specified sensitive elements of the environment do not occur, according to present knowledge” (Oslo Protocol, arts. 1(8) and 2(1); Gothenburg Protocol, arts. 1(12) and 2(1)). These critical loads result in fixed reductions of sulphur emissions for each state, which are listed in annexes to the protocols (Oslo Protocol, annex II; Gothenburg Protocol, annex II).

  • 2 Du points out that the estimated yearly global sulphur injections required for effective SAI deploy (...)

23To summarise, the CLRTAP sulphur protocols prohibit SAI if its emissions exceed the limits set by these protocols. However, the quantities of sulphur used for SAI would probably be relatively small compared to current overall sulphur emissions. Therefore, it is questionable whether the relevant protocols’ emissions ceilings would significantly restrict SAI (Brent 2017, 42-43; Lin 2013, 195-196). However, if tests or actual deployment of SAI were conducted on a global scale from the territory of a single country, the protocols’ limits could still be exceeded, which would prevent the use of the technology on this scale (Du 2016, 177-178; Reynolds 2019a, 99).2

2.2 Law Calling for Precaution

24A range of legal instruments that are potentially applicable to one or several of the case studies express the idea of exercising precaution, particularly concerning new technologies. Several instruments convey this message through references to the precautionary principle. In climate change law, the precautionary principle is notably contained in the UNFCCC:

The Parties should take precautionary measures to anticipate, prevent or minimise the causes of climate change and mitigate its adverse effects. Where there are threats of serious or irreversible damage, lack of full scientific certainty should not be used as a reason for postponing such measures, taking into account that policies and measures to deal with climate change should be cost-effective so as to ensure global benefits at the lowest possible cost. (art. 3(3))

25Proponents of SAI argue that the principle permits – or even requires – SAI and research into it by adopting a weak, permissive interpretation of the principle. They interpret the UNFCCC’s formulation of the precautionary principle as favouring the consideration of SAI as a climate change measure despite the uncertainty around the technology’s precise effects (Fleurke 2017, 207; Reynolds and Fleurke 2013, 106-107). They argue that the mention of cost-effectiveness in the UNFCCC’s precautionary principle encourages the use of comparatively cheap technologies such as SAI (Reynolds and Fleurke 2013, 107). Moreover, they point to several provisions in the UNFCCC that favour the development and diffusion of technology and thus might be seen as revealing a technology-friendly stance (arts. 4(1)(g)-(h), 4(3) and 4(5). See also Reynolds 2019a, 94).

26However, if interpreted in a stronger way, the precautionary principle would likely act as a barrier to SAI deployment, as the principle thus interpreted calls for caution when there are doubts about a measure’s effects (Bodansky 1996, 319; Lin 2013, 184). A strongly interpreted precautionary principle prefers reinforcing mitigation over using risky technologies (Bodle 2013, 459; Winter 2011, 281). This view is further comforted by the duty of UNFCCC parties to employ impact assessments or similar tools to minimise negative impacts on the environment or human health from measures addressing climate change (UNFCCC, art. 4(1)(f). See also Redgwell 2011, 184; Tedsen and Homann 2013, 97). Moreover, the precautionary principle has not been included in climate change law with a view to allowing measures against climate change despite the risks involved in these measures, but rather to allowing measures despite uncertainty about climate change itself (Bodle 2013, 459). While the Paris Agreement does not directly include the precautionary principle, its reference to the “best available science” may be read as restricting acceptable climate measures to proven, well-tested technologies (art. 4(1), preamble recital 4. See also Craik and Burns 2016, 7-8). The Agreement’s preamble further contains language signalling an aversion to risky climate policies. The preamble recognises that states may be negatively affected by “measures taken in response to [climate change]” and calls for the respect of human rights in taking and implementing such measures (Paris Agreement, preamble recitals 7 and 11. See also Rayfuse 2017, 74).

27Climate change law’s precautionary stance should be read in conjunction with the UNFCCC’s object and purpose. The objective of the UNFCCC is the “stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” (art. 2). This objective may particularly influence how climate change law sees SAI. As SAI does not affect GHG concentrations, this technology does not contribute to the UNFCCC’s objective (Reynolds 2014, 437). If the deployment of SAI drives states and economic actors to lessen mitigation efforts, the use of SAI might even be considered contrary to the objective of limiting GHG concentrations. Moreover, SAI itself might be considered a “dangerous anthropogenic interference” with the climate system and thus directly contrary to the Convention’s aim (Brent 2021, 280; Lin 2013, 183). It is true that the Convention’s objective only mentions dangerous anthropogenic interference stemming from high GHG concentrations in the atmosphere (Reynolds 2019a, 93). Nonetheless, employing SAI as a measure against climate change would arguably contradict the spirit and objective of the UNFCCC. In the same vein, it is questionable whether SAI would be consistent with the UNFCCC’s requirement to “protect the climate system”, as it does not aim at protecting current climate dynamics but would create a new climatic situation (art. 3(1). See also Lin 2013, 184).

28Similar arguments have been made about ocean fertilisation and the precautionary principle. Supporters point in particular to the fact that the UNFCCC’s formulation of the principle favours cost-effective measures (Güssow et al. 2010, 916). Moreover, they point out that the Convention’s primary objective is addressing climate change and may thus be read as favouring climate measures over avoiding the risks of environmental harm inherent in ocean fertilisation (Johansen 2020, 190-191; McGee, Brent, and Burns 2018, 74). As with SAI, a strong interpretation of the principle leads to the contrary conclusion. The potential side effects of ocean fertilisation block its use in a precautionary logic until there is enough scientific evidence that the benefits of ocean fertilisation outweigh its risks (Anderson and Peters 2016, 183; Lin 2013, 184; Rayfuse, Lawrence, and Gjerde 2008, 324-325). The latter view’s avoidance of risky technologies is also supported by the Paris Agreement’s call to use the “best available science” (art. 4(1) and preamble recital 4. See also Craik and Burns 2016, 7-8).

29Climate change law also calls for precaution when using adaptation measures such as artificial islands. As for mitigation, caution is warranted by the precautionary principle contained in the UNFCCC, particularly when read in conjunction with the Paris Agreement’s preambular statement that adverse effects of climate change measures are to be taken into account alongside the negative effects of climate change itself (UNFCCC, art. 3(3); Paris Agreement, preamble recital 7). Moreover, the Paris Agreement asks state parties to resort to the “best available science” on adaptation (art. 7(5)). This obligation might be interpreted as a call to caution against untested adaptation technologies whose risks are unknown. Finally, adaptation measures should consider the interests of vulnerable groups and, where appropriate, be integrated with socioeconomic and environmental policies (Paris Agreement, art. 7(5)). Applied to artificial islands, this rule requires that their construction should notably take into account the needs of underprivileged members of the society and the need to preserve traditional sources of income and cultural practices. The preamble’s call to respect human rights when taking measures to address climate change reinforces this need to consider socioeconomic factors in the choice of policies (Paris Agreement, preamble recital 11).

30Specifically for SAI, the law on the protection of the atmosphere similarly calls for caution. The CLRTAP’s core provisions are applicable to the technology even without exceeding the national emissions budgets determined by the protocols. Notably, parties have a general obligation to limit, reduce, and prevent air pollution (CLRTAP, art. 2. See also Reynolds 2018, 72). The Oslo Protocol reaffirms this general obligation specifically for sulphur emissions (art. 2(1)). Given this general obligation, SAI deployment by a CLRTAP party could be challenged for violating the general obligations and spirit of the Convention even if that party has not ratified any of the sulphur protocols (Bodansky 1996, 315-316). While the precautionary principle is not contained in the CLRTAP, the Oslo Protocol mentions precautionary measures in its preamble, and the Gothenburg Protocol’s preamble references the precautionary approach and Rio Principle 15 (Oslo Protocol, preamble recitals 3-4; Gothenburg Protocol, preamble recital 12). However, it is noteworthy that the critical loads approach of these protocols is somewhat at odds with the precautionary principle, as it permits a certain level of pollution, which is considered safe merely on the basis of current scientific knowledge (Dupuy and Viñuales 2018, 155). The LRTAP Convention also contains a precautionary obligation of information exchange and consultation with potentially affected states (arts. 4, 5, and 8). A “significant risk” of transboundary air pollution suffices to trigger an obligation by the party from which the pollution “originates or could originate” to consult with the affected party (CLRTAP, art. 5. See also Reynolds 2019a, 98).

31Technologies such as stratospheric aerosol injection and ocean fertilisation were also flagged as an area of concern early in the International Law Commission’s (ILC) discussions on the protection of the atmosphere (Murase 2014, paras. 73-74). The ILC’s work is intended to lead to the adoption of a set of draft guidelines that “do not seek to impose on current treaty regimes legal rules or legal principles not already contained therein” (Report of the ILC 2013, para. 168(d)). Guideline 7 explicitly deals with climate emergency technologies under the heading “intentional large-scale modification of the atmosphere”. In the latest version of the draft guidelines, adopted in 2021, guideline 7 reads as follows:

Activities aimed at intentional large-scale modification of the atmosphere should only be conducted with prudence and caution, and subject to any applicable rules of international law, including those relating to environmental impact assessment. (Protection of the Atmosphere: Texts and Titles of the Draft Guidelines and Preamble Adopted by the Drafting Committee on Second Reading 2021, guideline 7)

32The phrase “activities aimed at intentional large-scale modification of the atmosphere” is inspired by language in the ENMOD Convention, which prohibits using such activities as a military strategy (Report of the ILC 2018, 181-182). As the commentary to draft guideline 7 explicitly states, geoengineering is the prime example of the targeted activities (Report of the ILC 2018, 182). The commentary lists a wide array of technologies as falling under this term, including SRM techniques such as SAI and CDR techniques such as ocean fertilisation (Report of the ILC 2018, 182). The first draft of Guideline 7 by the special rapporteur explicitly used the term “geoengineering” in the guideline itself (Murase 2016, para. 91). Like other documents treating geoengineering without distinction, the guidelines have been criticised for this lack of nuance (Mayer 2019, 285-286). By implying that the same level of “prudence and caution” is necessary for widely different technologies, riskier measures might be downplayed while others unnecessarily hampered (Mayer 2019, 485-486).

33The guideline’s underlying philosophy is detailed in the commentary. First, the ILC recognises that atmosphere-modifying activities have a “significant potential” to prevent or moderate the negative impacts of droughts, storms, and other hazards but also entail the risk of unexpected impacts on the climate (Report of the ILC 2018, 182-183). Second, the drafters stress that they do not intend to stifle science and innovation but recognise that scientific innovation does not always have positive effects (Report of the ILC 2018, 183). Finally, the guideline does not aim at establishing a general rule allowing or prohibiting large-scale modification of the atmosphere, but at encouraging the development of rules to govern such activities (Report of the ILC 2018, 183).

34Guideline 7 demands that atmosphere modification is conducted “with prudence and caution”. This phrase is borrowed from language employed by the International Tribunal for the Law of the Sea (ITLOS) in several cases where a party had raised the precautionary principle (Report of the ILC 2018, 183; Mayer 2019, 485). The ITLOS’ use of this phrase has been considered an application of a precautionary approach or methodology while avoiding direct reference to the precautionary principle (Freestone 1999, 29-32). The use of this phrase by special rapporteur Murase can arguably be seen as an attempt to circumvent the limitations of his mandate that ban him from mentioning the precautionary principle (Mayer 2019, 485). The precautionary principle could not be included in the draft guidelines, as the ILC has decided to exclude the precautionary principle from the scope of its study on the protection of the atmosphere (Report of the ILC 2013, para. 168(a)). This “understanding” on the scope of the guidelines has been widely criticised as advance censorship that reduces the utility of the ILC’s whole study on the protection of the atmosphere (Mayer 2019, 462-466; Sand 2017, 205-207). The European Union and several states commented that the precautionary principle should be explicitly included in guideline 7 in spite of the understanding (Murase 2020, para. 65).

35The guideline’s mention of the rules of international law concerning environmental impact assessments (EIAs) is a further nod to the precautionary principle, as the former can be seen as a practical expression of the latter (Cançado Trindade 2015, 423-424). Nonetheless, the EU suggested a significantly more restrictive formulation, conditioning the deployment of any such technologies on a “positive opinion” by all potentially concerned states and the completion of a multinational EIA (Murase 2020, para. 65). Proponents of a precautionary stance further noted that, contrary to other guidelines in the ILC’s draft document, guideline 7 is drafted as a “should” requirement and argued for a change towards a “shall” requirement (Murase 2020, para. 67). Conversely, certain states and several ILC members considered that guideline 7 on large-scale modification of the atmosphere should not be included in the project, as the topic was too political, controversial, and evolving, and there was not yet sufficient state practice on the matter (Report of the ILC 2018, 183; Protection of the Atmosphere: Comments and Observations Received from Governments and International Organizations 2020, 33).

36For ocean fertilisation and artificial islands, the law of the sea’s environmental provisions also contain relevant precautionary language. The United Nations Convention on the Law of the Sea (UNCLOS) includes Part XII, which is dedicated to environmental protection. Its obligation to “protect and preserve the marine environment” is unqualified and applies to all activities in the oceans (UNCLOS, arts. 192-193. See also Verlaan 2007, 210). Notably, this obligation is explicitly applicable to marine scientific research, which might include research into ocean-based emergency technologies (UNCLOS, art. 240(d)).

37In particular for artificial islands, it is noteworthy that an arbitral tribunal has found China’s construction of artificial islands in the South China Sea to violate its obligation to protect and preserve the marine environment (The South China Sea Arbitration 2016, para. 983. See also Oral 2019, 435). Moreover, the tribunal found that China was obliged to cooperate with neighbouring states and conduct an EIA before constructing artificial islands (The South China Sea Arbitration 2016, paras. 986 and 988). In a similar case, the ITLOS underlined the pertinence of “prudence and caution” and obligations of information exchange and impact assessments in land reclamation activities, which present similar environmental and legal challenges as the construction of artificial islands (Land Reclamation by Singapore in and around the Straits of Johor 2003, para. 99. See also Stephens 2013, 286). Both these references demonstrate that the Convention’s environmental provisions reflect a precautionary approach despite the precautionary principle not being explicitly mentioned in the UNCLOS (Proelß 2018, 570-571). Furthermore, the UNCLOS’ duty to carry out an EIA before initiating potentially risky activities may be seen as an expression of the precautionary principle (UNCLOS, art. 206. See also Johansen 2020, 194; Scott 2018, 44). Finally, the ITLOS has explicitly confirmed the relevance of the precautionary principle in the framework of the UNCLOS (Responsibilities and obligations of States sponsoring persons and entities with respect to activities in the Area 2011, paras. 131-135. See also Johansen 2020, 194).

38In the case of ocean fertilisation, the UNCLOS’ more specific regulation of pollution might also be applicable, as the introduction of iron or other fertilising substances into the oceans may qualify as pollution if it negatively affects the marine environment (UNCLOS, arts. 194 and 196. See also Johansen 2020, 192-193; Freystone and Rayfuse 2008, 228-230; Scott 2018, 45). The precautionary principle is implicit in the Convention’s definition of pollution (Abate and Greenlee 2009, 574-575). This definition deems the introduction of substances which are “likely to result in […] deleterious effects” to constitute pollution, thus including cases where the effects of pollution cannot be determined with certainty (UNCLOS, art. 1(1)(4) (emphasis added). See also Abate and Greenlee 2009).

39Of particular importance for ocean fertilisation might be article 195 of the UNCLOS (Verlaan 2009, 458). This provision provides that states’ pollution control measures shall not “transfer, directly or indirectly, damage or hazards from one area to another or transform one type of pollution into another” (UNCLOS, art. 195). Ocean fertilisation might be argued to constitute a breach of this provision, as it addresses the environmental problems created by greenhouse gases by introducing other potential pollutants into the oceans. Therefore, proponents of ocean fertilisation would need to prove that their projects do not simply turn climate change into marine pollution and thus negatively affect marine life by transferring hazards from the atmosphere to the ocean (Verlaan 2009, 458).

40As a specific form of pollution, dumping is subject to more specific obligations (UNCLOS, art. 210). The parties to UNCLOS are obliged to cooperate in creating global regulation of dumping and to ensure that their national legislation is at least as effective as these global rules (UNCLOS, arts. 210(4) and 210(6). These global rules are provided by the London Convention and the London Protocol (Rayfuse, Lawrence, and Gjerde 2008, 311). The question of whether ocean fertilisation constitutes dumping has been addressed in the framework of these treaties, which will be discussed below.

2.3 Law Imposing a Precautionary Moratorium

41Faced with a risky new technology, one possible precautionary response is to impose a moratorium. Trouwborst describes the moratorium as “the precautionary measure par excellence” (Trouwborst 2007, 189). A moratorium often has the effect of reversing the burden of proof in the sense that the moratorium’s ban on a risky activity is lifted only once its proponents prove its harmlessness (Trouwborst 2007, 192).

42The London Convention (LC) and the London Protocol (LP) regulate ocean fertilisation in a precautionary manner, including by imposing a moratorium on certain activities. The two London treaties have been the most active international forum explicitly discussing and regulating ocean fertilisation, thus “claiming ownership” of the issue (Eick 2010, 366-368). These treaties aim at regulating the dumping of waste and similar material at sea to prevent pollution of the marine environment (London Convention, art. I; London Protocol, art. 2). The LP is designed to eventually replace the LC, as it constitutes a “modern version” of the LC, which is based on the precautionary principle (Ginzky 2010, 63-64). In the meantime, the LC and LP interact in a “two instruments – one family” approach, with parallel meetings of the parties and some resolutions being adopted simultaneously for both instruments (Verlaan 2009, 453). Between states parties to both instruments, the LP supersedes the LC (London Protocol, art. 23).

43In parallel, the parties to the CBD have also adopted guidance on ocean fertilisation, with numerous cross-references between this regime and the LC/LP. The CBD’s participation in ocean fertilisation regulation is significant in particular because of its near-universal ratification. Thus, the CBD could contribute to closing gaps in the jurisdiction of the London instruments and notably the scarcely ratified London Protocol (Ginzky and Frost 2014, 93). As the CBD applies not only on the territory of its parties but also to activities under their jurisdiction or control beyond the limits of national jurisdiction, this signifies that regulation under the CBD has the potential to reach almost all ocean fertilisation activities globally (CBD, art. 4. See also Güssow et al. 2010, 915). Materially, ocean fertilisation falls under the scope of the CBD because ocean fertilisation activities might impact biological diversity. Under the CBD, parties are obliged to identify activities that are “likely to have significant adverse impacts on the conservation and sustainable use of biological diversity” and to regulate such activities (CBD, arts. 7(c) and 8(l). See also Eick 2010, 362). In light of ocean fertilisation’s risks for harm to marine ecosystems, the CBD can thus be applied to ocean fertilisation (Güssow et al. 2010, 915).

44Before the LC and LP governing bodies actively engaged with ocean fertilisation, there was significant debate on whether the technology falls under the treaties’ definition of dumping (Eick 2010, 365-366; Freestone and Rayfuse 2008, 229-230; Ginzky 2010, 64-65; Rayfuse, Lawrence, and Gjerde 2008, 312-317). Besides these early discussions, engagement with ocean fertilisation has also been facilitated by the fact that amendments to the LP permitting and regulating CO2 sequestration in the seabed were adopted previously (LP Resolution LP.1(1); LP Resolution LP.3(4). See also VanderZwaag and Daniel 2009, 532-534). Direct engagement with ocean fertilisation started in 2007 when the LC and LP Scientific Groups issued a “statement of concern” about ocean fertilisation (Report of the 30th Meeting of the Scientific Group of the LC 2007, paras. 2.23-2.26). The issue was brought to the Groups’ attention by NGO observers and the United States, who reacted to a US-based company’s planned large-scale ocean fertilisation experiment (Greenpeace International 2007; International Union for Conservation of Nature and Natural Resources 2007; United States of America 2007). The statement underlined the risks for negative impacts on the marine environment and recommended careful examination of such projects (Report of the 30th Meeting of the Scientific Group of the LC 2007, para. 2.25).

45In the same year, the meetings of the contracting parties to the LC and LP endorsed the Scientific Groups’ statement of concern and recognised that the LC and LP were competent to address ocean fertilisation (Report of the 29th Consultative Meeting 2007, para. 4.23). While the scientific groups’ statement was limited to ocean iron fertilisation, the contracting parties’ statement broadened the scope to reference all types of ocean fertilisation (Report of the 29th Consultative Meeting 2007, para. 4.23.2. See also VanderZwaag and Daniel 2009, 536-537). The LC and LP’s competence to regulate the issue was based on the two treaties’ general objective to “protect and preserve the marine environment from all sources [of pollution]” (Report of the 29th Consultative Meeting 2007, para. 4.23.3). While agreeing to further consider the issue in the future, the meeting urged all states to consider proposals for large-scale ocean fertilisation operations with the “utmost caution” and noted that such operations were not justified at that time, given the limited state of knowledge (Report of the 29th Consultative Meeting 2007, para. 4.23.4-4.23.5). In December 2007, the United Nations General Assembly (UNGA) welcomed the decision of the LC and LP parties and encouraged further study of ocean iron fertilisation in its resolution 62/215 on the oceans and the law of the sea (UNGA Resolution 62/215, paras. 97-98).

46The CBD’s Conference of the Parties (COP) first manifested its engagement with ocean fertilisation in 2008 in its Decision IX/16 on biodiversity and climate change. One part of this resolution is dedicated to ocean fertilisation (CBD Decision IX/16, part C). Therein, the CBD Parties welcome the LC/LP’s 2007 statement and request parties to follow its guidance (CBD Decision IX/16, part C.2). Furthermore, the decision requests governments “in accordance with the precautionary approach, to ensure that ocean fertilization activities do not take place until there is an adequate scientific basis on which to justify such activities, including assessing associated risks, and a global, transparent and effective control and regulatory mechanism is in place for these activities” (CBD Decision IX/16, part C.4). Simultaneously, the decision explicitly allows small-scale scientific studies in coastal waters for which it requires a “thorough prior assessment” of environmental impacts and forbids the sale of carbon credits or other commercial use (CBD Decision IX/16, part C.4).

47In the same year, the parties of the LC and LP adopted Resolution LC-LP.1, exclusively devoted to ocean fertilisation. The resolution was not intended to be legally binding but can be referred to for purposes of interpretation (Ginzky 2010, 66; Schlacke and Kenzler 2009, 755-756. See also London Protocol, art. 18(1)(7)). The resolution reaffirms that ocean fertilisation falls under the Convention’s and Protocol’s scope (LC/LP Resolution LC-LP.1, para. 1). This is the first resolution containing a definition of ocean fertilisation, defining it as “any activity undertaken by humans with the princip[al] intention of stimulating primary productivity in the oceans” (LC/LP Resolution LC-LP.1, para. 2). Importantly, the LC/LP parties agreed that such activities should be considered contrary to the aims of the Convention and Protocol (LC/LP Resolution LC-LP.1, para. 8). This means that ocean fertilisation is considered dumping and thus covered by the treaties because the LC and LP consider the “placement of matter for a purpose other than the mere disposal thereof” to constitute dumping if it is contrary to the treaties’ aims (London Convention, art. III(1)(b)(ii); London Protocol, art. 1(4)(2)(2)). However, if a specific ocean fertilisation project constitutes “legitimate scientific research”, it is not considered contrary to the aims of the treaties and is thus permitted (LC/LP Resolution LC-LP.1, paras. 3-8).

48The 2008 resolutions by the CBD and LC/LP have been described as (non-binding) moratoria on non-research ocean fertilisation activities (Ginzky and Frost 2014, 83 and 93). At the adoption of the CBD decision, Ecuador’s representative expressed his regret that the term moratorium was not included in the text of the decision, but he underlined that the text should nonetheless be interpreted as calling for a moratorium on ocean fertilisation (Report of the CoP to the CBD 2008, para. 250). However, both resolutions leave the door open for ocean fertilisation in the future by qualifying their statements about the insufficiency of scientific knowledge and their prohibition with the words “currently” and “until there is an adequate scientific basis on which to justify such activities” (LC/LP Resolution LC-LP.1, preamble recital 6 and para. 8; CBD Decision X/33, para. 8(w). See also Verlaan 2009, 457).

49Despite the many parallels, there are certain discrepancies between the CBD decision and the LC/LP resolution, the main difference being that only the CBD decision limits permissible research to coastal waters and thus prohibits experiments on the high seas (Schlacke and Kenzler 2009, 756-758). This additional limitation in the CBD instrument may facilitate parties’ control over experiments but arguably renders any useful experimentation impossible, as coastal waters are not suitable for ocean fertilisation experiments because they already contain large quantities of nutrients from land-based sources (Schlacke and Kenzler 2009, 757; Branson 2014, 176).

50In 2010, the meeting of the parties to the LC and LP adopted Resolution LC-LP.2 containing an assessment framework for ocean fertilisation research projects. The framework is designed to assist state parties in determining whether a proposed ocean fertilisation project constitutes “legitimate scientific research” and is thus consistent with the Convention’s and Protocol’s aims (LC/LP Resolution LC-LP.2, 2-3. See also LC/LP Resolution LC-LP.1, paras. 4-5). The framework outlines the steps to determine whether a project has scientific attributes, carry out an environmental assessment, and decide based on the environmental assessment. The precautionary approach plays a central role in the framework. The risk management stage of the assessment is “based on a precautionary approach”, and such an approach has to be “followed” (LC/LP Assessment Framework 2010, paras. and 3.6.1). Moreover, at the decision stage, the precautionary approach has to be “tak[en] into account” and obliges states to reject project proposals when “the risks and/or uncertainties are so high as to be deemed unacceptable, with respect to the protection of the marine environment” (LC/LP Assessment Framework 2010, para. 4.3). In this assessment, uncertainties have to be identified and taken into account (LC/LP Assessment Framework 2010, para. 3.5.13). Moreover, potentially directly impacted states should be consulted and their consent requested (LC/LP Assessment Framework 2010, paras. 3.6.3 and 4.2). However, this consultation requirement might be less meaningful than one might expect, given the likely conduct of ocean fertilisation activities in the high seas.

51With its obligations to reject high-risk proposals, the Assessment Framework’s approach can be considered highly cautious (Scott 2018, 50). However, many of the prescriptions of the assessment framework are worded in a way that gives the deciding state party considerable latitude, frequently using the word “should” (Eick 2010, 370). Moreover, the framework seems to encourage a balancing exercise between scientific benefits and environmental risks, thus deviating from a strictly precautionary approach (LC/LP Assessment Framework 2010, paras.3.6.1 and 4.1). Finally, the development of an assessment framework as such might ultimately facilitate the deployment of the technology. Such a document, by providing the conditions under which ocean fertilisation should be allowed, ends the debate on whether it should be permitted at all (Scott 2013, 116).

52At the same time as the LC and LP meetings adopted the assessment framework for ocean fertilisation, the COP of the CBD adopted another decision on biodiversity and climate change. Decision X/33 mirrors the language from the previous Decision IX/16 on ocean fertilisation but extends its scope to all types of “climate-related geoengineering”, which clearly includes ocean fertilisation but is otherwise not defined (CBD Decision X/33, para. 8(w)). It invites states to ensure that no such activities take place until there is a sufficient scientific basis to justify them, except for small-scale research (CBD Decision X/33, para. 8(w)). The decision resolves the inconsistency with the LC/LP concerning the criteria for permissible research, as it drops the reference to the coastal proximity of experiments. Like the 2008 CBD Decision on ocean fertilisation, CBD Decision X/33 refers to the precautionary approach (CBD Decision X/33, para. 8(w)).

53The subsequent CBD Decision XI/20 additionally notes that the precautionary principle and other rules of customary international law alone are not sufficient for global regulation of geoengineering (CBD Decision XI/20, para. 11). This decision also explicitly states the primacy of emission reductions and sinks recognised by the UNFCCC over any geoengineering technology (CBD Decision XI/20, para. 4). The CBD’s latest decision on ocean fertilisation from 2016 does not add any new substantive content but reaffirms the earlier decisions and notes the lack of response by CBD parties to earlier decisions’ calls for information exchange on the issue (CBD Decision XIII/14).

  • 3 Currently, only six parties have accepted the amendment. See Overview of the Status of the London C (...)

54As the LC/LP and CBD resolutions are not binding, the parties to the LP decided to go further in 2013. They adopted an amendment to the Protocol to include ocean fertilisation and “other marine geoengineering activities” in the text of the treaty (Ginzky and Frost 2014, 84). The amendment is not yet in force, as less than two-thirds of the parties to the Protocol have accepted it.3 The amendment adds a definition of “marine geoengineering” and a new provision regulating such activities to the LP. The new Article 6bis prohibits those marine geoengineering activities that are listed in the newly created Annex 4 except where the annex allows a specific activity subject to a permit (LP Resolution LP4.8, arts. 1(5bis) and 6bis).

55The amendment to the LP defines marine geoengineering as follows:

“Marine geoengineering” means a deliberate intervention in the marine environment to manipulate natural processes, including to counteract anthropogenic climate change and/or its impacts, and that has the potential to result in deleterious effects, especially where those effects may be widespread, long lasting or severe. (LP Resolution LP4.8, art. 1(5bis))

56The second part of this definition of marine geoengineering closely follows the definition of prohibited hostile environmental modification in the ENMOD Convention (ENMOD Convention, art. I(1). See also Ginzky and Frost 2014, 86). However, while the definition of marine geoengineering does not require the activities to be hostile, an additional qualifier limiting the scope to activities with “deleterious” effects on the marine environment has been added, which stems from the LP’s definition of pollution (Ginzky and Frost 2014, 86; London Protocol, art. 1(10)). The definition clarifies that marine geoengineering includes, but is not limited to, interventions designed to counteract climate change or its effects (LP Resolution LP.4(8), art. 1(5bis)). This precision partly closes a potential loophole, as some ocean fertilisation projects have claimed to aim at increasing fish growth rather than carbon sequestration (Ginzky and Frost 2014, 86). However, the annex’s definition of ocean fertilisation explicitly excludes “conventional aquaculture, or mariculture, or the creation of artificial reefs” (LP Resolution LP.4(8), annex 4, para. 1.1). Thus, a loophole might still exist, as the stimulation of fish growth could be labelled mariculture, which would allow ocean fertilisation to be disguised as a fisheries-related activity (Jeffries 2017, 29-30).

57Currently, the only activity listed in Annex 4, and thus covered by the amendment, is ocean fertilisation (LP Resolution LP.4(8), annex 4). The parties have deliberately chosen an approach allowing them to add new geoengineering techniques following the simplified procedure for amendments of annexes, which does not require ratification by state parties (Report of the 35th Consultative Meeting 2013, paras. 4.3 and 4.12.1; London Protocol, art. 22). The same flexibility applies to amending the current coverage of ocean fertilisation, making it relatively easy for parties to abolish the ban on deployment (Ginzky and Frost 2014, 92). This flexibility might facilitate the usage of ocean fertilisation as an emergency solution, as the annex can be adapted in a relatively short period of time. However, the legal framework provides some safeguards against hastened deployment in the event of emergency, as a two-thirds majority of voting parties is necessary to amend the annex (London Protocol, art. 22(2)).

58By regulating only activities listed in the newly created annex, the amendment deviates from the LP’s “reverse list” approach under which all dumping of waste or other matter not listed in Annex 1 is prohibited, while the dumping of listed substances can be allowed through a permit (London Protocol, art. 4(1)). This reverse list approach has been described as a “full embracement” of the precautionary principle (Johansen 2020, 195). Indeed, this approach involves a reversal of the burden of proof by ensuring that dumping of new, unknown substances is only allowed after parties have examined the scientific evidence on the potential risks and two-thirds of them have approved its addition to the annex (London Protocol, art. 22(2). See also Rayfuse, Lawrence, and Gjerde 2008, 310). Thus, the reverse list ensures that the dumping of substances remains prohibited as long as its effects remain uncertain. In contrast, the new article 6bis prohibits only the activities listed in the new Annex 4, following a “positive list” approach rather than banning all activities that fall under the definition of marine geoengineering (Report of the 35th Consultative Meeting, para. 4.4; Ginzky and Frost 2014, 87). This departure from the LP’s overall approach was justified by the fact that the effects and risks of activities other than ocean fertilisation that could fall under the definition were not yet sufficiently known to determine the appropriate levels of control (Ginzky and Frost 2014, 87). Consistent application of the precautionary principle would require also submitting new, yet unknown, forms of geoengineering to the new control regime.

  • 4 According to Ginzky and Frost 2014, 88, the qualifier ‘as far as practicable’ only applies to the o (...)

59The amendment allows some sub-categories of marine geoengineering to be authorised by states. For the activities that Annex 4 allows states to permit, the proposed project has to be assessed following the general framework introduced in Annex 5 and considering specific frameworks (LP Resolution LP.4(8), art. 6bis(2) and annex 5). Ocean fertilisation, the only activity included in Annex 4 to date, is generally prohibited, but legitimate scientific research is permitted, subject to a permit from the state party in which the vessel is loaded (LP Resolution LP.4(8), annex 4, para 1.3 and art. 9(2)). When adopting the amendment, the parties clarified that the above-discussed 2010 assessment framework would serve as the appropriate specific framework for ocean fertilisation (LP Resolution LP.4(8), para. 3). The assessment is to be made exclusively by the concerned state party, in consultation with other potentially impacted states and taking into account advice from international experts (LP Resolution LP.4(8), annex 5, paras. 10-12). Following the assessment, parties can grant a permit if the assessment shows that the activity is not contrary to the Protocol’s aims and that pollution from the activity is “as far as practicable, prevented or reduced to a minimum” (LP Resolution LP.4(8), art. 6bis(2).4

60The general assessment framework in Annex 5 underlines that activities should not be permitted if the available information is not sufficient to assess their likely effects (LP Resolution LP.4(8), paras. 6 and 18). This means that the burden of proof lies with the proponents of the activity (Ginzky and Frost 2014, 89). Furthermore, article 3(1) of the Protocol, obliging parties to use a precautionary approach, is amended to explicitly apply to marine geoengineering covered by the amendment (LP Resolution LP.4(8), art. 3(1). See also Ginzky and Frost 2014, 91). Nonetheless, as discussed above, some elements of the assessment framework’s implementation of precaution remain problematic.

2.4 Ambiguous Law

61Some legal instruments are relevant and applicable to emergency technologies but are ambiguous regarding their stance towards them. Often, the rules and language contained in these instruments can be used both to argue in favour of and against the use of such technologies or can have both facilitating and hindering effects. Such dynamics can be found in climate change law relating to SAI and the law applicable to artificial islands. In a certain way, many of the previously studied laws calling for precaution could be argued to fall into this category, as some authors do not see precaution as calling for caution in using such technologies but rather as militating for using such technologies as a precaution against climate change despite the doubts about their effectiveness and harmlessness (Fleurke 2017, 207; Reynolds and Fleurke 2013, 106-107).

62Climate change law is particularly ambiguous towards SAI, as its instruments do not contain any explicit rules prohibiting or addressing this technology, which does not neatly fall into the categories of mitigation and adaptation (Brent 2017, 40). Furthermore, discussions at the UNFCCC COPs have not directly addressed the technology; UNFCCC negotiators have to date avoided addressing this highly politicised “elephant in the room” (Craik and Burns 2016, 5-6; Horton, Keith, and Honegger 2016, 5). Consequently, it has, for instance, not been clarified or discussed whether the Paris Agreement allows states to include SAI projects in their climate efforts (Craik and Burns 2016, 6; Reynolds 2019a, 96). The Paris Agreement is based on nationally determined contributions (NDCs) submitted by each party. Article 3 is open about the types of contributions included in NDCs, which might open the possibility for states interested in SAI to include it in their NDC submission. However, article 4 spells out that sinks and removals should balance greenhouse gas emissions by the second half of the century (art. 4(1)). Given this focus on GHG emissions and concentrations, it is doubtful if SAI could be considered a mitigation measure to be included in an NDC (Craik and Burns 2016, 8). Moreover, SAI hardly qualifies as adaptation or as a measure addressing loss and damage (UNFCCC, arts. 7-8). The provisions on the two latter issues focus on measures that address the effects of climate change rather than intervening directly in climate systems as SAI would (Craik and Burns 2016, 9-10).

63Concerning artificial islands, the rules of law that would be relevant for recognising them as a replacement for disappearing natural territory, namely the law of the sea and the rules on statehood, are similarly ambiguous. The circumstances under which artificial islands can be erected are relatively clear, as coastal states have the right to construct them in their exclusive economic zone and on the continental shelf and to exercise exclusive jurisdiction over them, as long as the artificial islands do not interfere with maritime navigation (UNCLOS, arts. 60 and 80). On the high seas, any state is permitted to construct artificial islands (UNCLOS, art. 87(1)(d)). The main ambiguity lies in whether artificially created islands could be considered state territory and thus potentially be used to replace disappearing state territory. The answer to this question determines whether a state could continue to exist on an artificial island if its natural territory is lost to sea-level rise, as an inhabitable territory is arguably indispensable for a state’s continued existence (Bergmann 2016, 132-133).

64The UNCLOS does not explicitly define artificial islands, but they can be defined in opposition to the Convention’s definition of “real” islands. An island is defined as “a naturally formed area of land, surrounded by water, which is above water at high tide” (UNCLOS, art. 121(1). See also Aurescu and Oral 2020, para. 218(e); Oude Elferink 2013, paras. 1-3). Moreover, in order to justify claims to maritime zones beyond the territorial sea, islands must be able to “sustain human habitation or economic life of their own” (UNCLOS, art. 121(3). See also Soete 2021, 247-250). In principle, the UNCLOS excludes the possibility of using an artificial island to claim jurisdiction over maritime zones, as such islands explicitly do not have any territorial sea or other maritime zones of their own (UNCLOS, art. 60(8)). However, artificial islands could still constitute state territory. The UNCLOS has arguably only abolished the possibility of artificial islands generating maritime zones, while it has not changed the previously established principle that artificial islands can be considered state territory to the same degree as naturally formed ones (Grote Stoutenburg 2015, 171-172).

65If an artificial island is a floating structure rather than an island created by building up materials on the seafloor, the current legal regime is less likely to accept a claim of this structure constituting state territory. This is because floating islands presumably have to be considered installations or structures under the UNCLOS (art. 60(1)(b)). While the regime for installations and structures is virtually the same as for artificial islands under the UNCLOS, the argument for recognising such a structure as state territory is weakened under the laws on statehood, as structures were never considered part of state territory (Saunders 2019, 649). Nevertheless, other states might also accept floating islands as sinking states’ replacement territory in light of considerations of fairness and equity (Grote Stoutenburg 2015, 175).

66The customary rules on statehood are generally similarly ambiguous as the law of the sea, as there are no precedents for a state disappearing without a successor state taking its place (Bergmann 2016, 127-128; Crawford 2007, 715). However, there is a strong presumption towards the continuity of states even if they lose large parts of their territory or an effective government (Crawford 2007, 700-701). Thus, the reaction of other states towards an island state’s claim to continuance based on an artificial island would presumably be crucial, as recognition can serve as the decisive factor in cases where statehood is in doubt (Crawford 2007, 93; Grote Stoutenburg 2015, 296). If the sea-level rise leading to the disappearance of the state territory was found to stem from the violation of a jus cogens norm, other states would even be obliged to not recognise the state’s disappearance (Soete 2021, 156). Even in the absence of such an obligation, it is imaginable that other states would also recognise continuing claims to maritime zones, as the rule limiting such claims has only been inserted in the UNCLOS to avoid states expanding their existing maritime zones; the use of artificial islands to simply maintain claims has not been contemplated by the UNCLOS’ drafters (Grote Stoutenburg 2015, 172-173; Soete 2021, 250). However, there are legitimate doubts as to whether the presumption of continuity and recognition by other states would suffice to remedy the absence of any natural territory (Bergmann 2016, 129-131).

67The ambiguity of both regimes is confirmed by the fact that several authors propose amendments to the UNCLOS or a new treaty on artificial islands to create legal stability for sinking island states that plan to use artificial islands to avoid their looming disappearance (Gagain 2012, 107; Papadakis 1977, 103-104). Alternatively, the conclusion of ad-hoc treaties with concerned states, recognising their continued existence, is proposed to overcome legal ambiguities (Rayfuse and Crawford 2011, 11). Some degree of clarification could also come from the ILC, which is currently conducting work on different issues concerning sea-level rise and international law. In the first issues paper by the study group, the legal status of artificial islands is among the legal questions that are preliminarily examined (Aurescu and Oral 2020, paras. 211-218). A report on questions relating to sea-level rise and statehood is to be elaborated by the relevant study group in the near future (Aurescu and Oral 2020, para. 225).

2.5 Encouraging Law

68Some provisions in climate change law may be read as encouraging emergency technologies as mitigation or adaptation measures. Since the adoption of the UNFCCC in 1992, the objectives of climate change law have evolved with political developments, advances in climate modelling, and new technological promises (McLaren and Markusson 2020). The Paris Agreement, as the most recent climate agreement, features a temperature limit as its primary objective (art. 2(1)(a). See also McLaren and Markusson 2020, 395). This focus on temperature as the main indicator of success may facilitate the use of SAI and ocean fertilisation as climate policies. While the UNFCCC’s focus on greenhouse gases excluded SAI from its scope of application, the Paris Agreement seems to be more welcoming for measures like SAI that solely address temperature rise without affecting GHG concentrations (Horton, Keith, and Honegger 2016, 5). Moreover, setting the target at “well below 2 °C” (Paris Agreement, art. 2(1)(a) might make SAI or similar technologies appear necessary to achieve the target, as mitigation measures might not suffice (Craik and Burns 2016, 5; Horton, Keith, and Honegger 2016, 5. See also Nationally determined contributions under the Paris Agreement 2021). For ocean fertilisation or other CDR technologies, the Paris Agreement’s temperature-focused objective might similarly facilitate deployment. As most models for limiting temperature rise to 2 °C or 1.5 °C assume the use of carbon removals, ocean fertilisation is likely to be proposed as a means to achieve the Paris Agreement’s objective (Craik and Burns 2016, 4-6; Horton, Keith, and Honegger 2016, 3; McLaren and Markusson 2020, 395).

  • 5 Submission of long-term climate strategies is mandated by Paris Agreement, art. 4(19).

69Climate change law is more encouraging towards ocean fertilisation than SAI. Ocean fertilisation is more similar to traditional mitigation measures than SAI and is thus more likely to be recognised as a climate change measure under the UNFCCC and the Paris Agreement (Tedsen and Homann 2013, 95). In particular, removing carbon from the atmosphere through ocean fertilisation corresponds to the UNFCCC’s overall objective of stabilising GHG concentrations in the atmosphere (art. 2). As states have considerable freedom in determining their climate measures through NDCs, the inclusion of ocean fertilisation in an NDC is the most probable way for ocean fertilisation to find its way into the Paris Agreement (Craik and Burns 2016, 6; Johansen 2020, 190). While ocean fertilisation has not yet been included in any NDC, Switzerland’s long-term climate strategy mentions ocean fertilisation as one of several potential negative emissions technologies to be used to compensate emissions that are not eliminated until 2050 (Federal Council 2021, 48).5 Such inclusion in long-term strategies, and eventually in NDCs, is made possible by the UNFCCC’s and Paris Agreement’s recognition of carbon sinks as mitigation measures (UNFCCC, art. 4(1)(d); Paris Agreement, arts. 4(1) and 5(1). See also Fleurke 2017, 207-208). The UNFCCC defines a sink as “any process, activity or mechanism which removes a greenhouse gas, an aerosol or a precursor of a greenhouse gas from the atmosphere”, thus not excluding artificial enhancement of sinks such as ocean fertilisation (UNFCCC, art. 1(8). See also Craik and Burns 2016, 6). Moreover, the UNFCCC explicitly includes the oceans as a possible carbon sink or reservoir (art. 4(1)(d). See also Reynolds 2014, 437). Article 5 of the Paris Agreement calls on states to “enhance” GHG sinks and reservoirs (art. 5(1)). This provision may be argued to encourage ocean fertilisation, as the technology’s aim is precisely to enhance the naturally occurring carbon cycle of the oceans (Craik and Burns 2016, 7; Johansen 2020, 190).

70The UNFCCC regime might also encourage the construction of artificial islands, as this pursuit might be recognised as an adaptation measure. The UNFCCC and the Paris Agreement require states to take measures for adapting to climate change (UNFCCC, art. 4(1)(b); Paris Agreement, art. 7). Under the Paris Agreement, measures to counter sea-level rise might alternatively also fall under its provisions on preventing loss and damage from climate change, which cover “slow onset events” such as sea-level rise (Paris Agreement, art. 8(4)(c). See also Oral 2019, 423). In line with the Paris Agreement’s bottom-up approach, each state party is to determine its priorities in adapting to climate change and its effects (arts. 7(5) and 7(9)). These priorities, along with the state’s planned actions and its needs for support by other states, are to be submitted through regular adaptation communications (Paris Agreement, art. 7(10)). While only a few states have already submitted such documents, future communications, notably from small island states, might include artificial islands as an adaptation measure. In the current adaptation communication by the Marshall Islands, the only island state to have submitted its plans to date, protecting infrastructure and maritime borders from sea-level rise is among the identified areas of adaptation needs (Republic of Marshall Islands 2020, 28-30).

2.6 Law Facilitating Funding

71Besides encouraging ocean fertilisation as a mitigation measure and artificial islands as an adaptation measure in the abstract, climate change law might also facilitate the funding of projects involving these technologies. The same is true for SAI if it comes to be recognised as a mitigation or adaptation measure. As a sink-enhancing measure, ocean fertilisation is particularly likely to profit from the climate change regime’s tools for securing the necessary financial resources. Developed states in the Paris Agreement commit to assist developing states financially in their mitigation and adaptation efforts (arts. 9(1) and 9(4)). If ocean fertilisation is read as a sink-enhancing mitigation measure, this obligation might provide financial means for ocean fertilisation projects, notably in developing countries (Chichilnisky and Bal 2016, 263). Such finance could be provided bilaterally or through the multilateral Green Climate Fund (Chichilnisky and Bal 2016, 263. See also UNFCCC, art. 11; Paris Agreement, art. 9(8)).

72Moreover, the climate regime might facilitate ocean fertilisation by allowing its carbon sequestration to be monetised through market mechanisms. Such mechanisms would presumably be a critical element in ocean fertilisation development, as they would make it attractive for private companies to invest in these expensive technologies (Craik and Burns 2016, 8). However, offsets stemming from ocean fertilisation were not accepted under the market mechanisms of the Kyoto Protocol, and none of the currently existing regulated carbon markets accept their trade (Bertram 2010, 1135; Freestone and Rayfuse 2008, 231; Johansen 2020, 189). Under the Paris Agreement, negotiations on the future rules of the Sustainable Development Mechanism will need to determine whether ocean fertilisation is eligible for creating tradeable carbon credits (Paris Agreement, arts. 6(4) and 6(7). See also Craik and Burns 2016, 8-9). Adopting regulation for ocean fertilisation that would allow its inclusion in carbon markets might be more challenging than for other methods of creating carbon credits, mainly because the additionality and permanence of carbon sequestration are difficult to demonstrate in ocean fertilisation projects (Bertram 2010, 1135-1138; Lin 2013, 185).

73As for mitigation, developed country parties are obliged to provide financial resources for adaptation measures to developing states, which might include the financing of artificial islands (Paris Agreement, art. 9(1)). As financial assistance has to take into account “country-driven strategies” and the priorities and needs of small island states, the inclusion of artificial islands in these states’ adaptation plans would facilitate their financing through developed states’ resources (Paris Agreement, art. 9(5)). The UNFCCC regime also provides institutional mechanisms for cooperation and support for adaptation measures. Notably, the Adaptation Committee and the Warsaw International Mechanism for Loss and Damage provide forums for cooperation on measures that might include artificial islands (Paris Agreement, arts. 7(7) and 8(2). See also Oral 2019, 421-422). Moreover, the Green Climate Fund as the UNFCCC’s financial mechanism can provide financial resources for adaptation measures, potentially including the construction of artificial islands (UNFCCC, art. 11; Paris Agreement, art. 9(8). See also Oral 2019, 422). Finally, the development and use of all types of technology are encouraged through the UNFCCC’s Technology Mechanism that provides technical assistance for developing and transferring climate technology (Paris Agreement, art. 10(3)-(5)).

2.7 Law Facilitating Research

74As the three emergency technologies and their effects and side effects are not yet well studied, law facilitating such research could have a significant impact on the prospects of future deployment. Such facilitation may notably be found in the law of the sea, climate change law and the LRTAP Convention.

75The law of the sea contains numerous provisions protecting and encouraging marine scientific research, which might facilitate research into ocean fertilisation and artificial islands. The UNCLOS protects marine scientific research as one of the freedoms of the high sea (arts. 87(1) and 238). Therefore, all states have the right to conduct research activities on the high seas (UNCLOS, arts. 87(1) and 238). This protection as a freedom of the high seas is particularly relevant for ocean fertilisation, which is likely to occur in this part of the ocean.

76In addition to recognising research as a freedom of the high seas, the UNCLOS supports marine scientific research through Part XIII. Among other things, Part XIII obliges states to promote such research and create favourable conditions for it (UNCLOS, arts. 239 and 243). However, even if research into ocean fertilisation or artificial islands falls under the freedom of scientific research, this freedom is limited, as it has to be exercised with due regard to other states’ interests (UNCLOS, arts. 87(2), 238 and 240(d)). Thus, for example, ocean fertilisation would not be permitted if it interfered with fisheries or shipping (Scott 2018, 43). Moreover, all other provisions of the UNCLOS have to be complied with when conducting marine research, notably including environmental provisions (UNCLOS, art. 240(d). See also Verlaan 2007, 211).

77Specific provisions in the LRTAP Convention might similarly be interpreted as facilitating or encouraging research and development of SAI technologies. For example, Reynolds argues that GHG emissions should be considered air pollution under the Convention’s definition, and SAI consequently seen as a technology to manage pollution (Reynolds 2019a, 98-99). Specifically, article 7, which mandates research into measures for reducing pollution and for other environmental objectives, could be interpreted as encouraging research into SAI technologies to combat the effects of pollution by GHGs (Reynolds 2019a, 98-99. See CLRTAP, art. 7(e)). Similarly, the reference in article 6 to “the best available technology which is economically feasible” (BATEF) for air quality management systems might be understood as demanding the inclusion of SAI into such systems (Reynolds 2019a, 99. See CLRTAP, art. 6).

78However, this interpretation of the LRTAP Convention as encouraging research into emergency climate technologies seems to contradict the overall logic and objectives of the Convention. First, such a reading contradicts the Convention’s preference for preventing air pollution over treating its effects (Brent 2021, 280-281). The Convention’s stated objective is the limitation, reduction, and prevention of air pollution (CLRTAP, art. 2). Therefore, the obligations of the Convention and its protocols all focus on reducing emissions, not on treating their effects. In light of this objective, SAI could in no way be considered a measure for reducing GHG emissions under the LRTAP Convention, as it would increase emissions of pollutants rather than reduce them. In other words, the LRTAP Convention does not favour addressing pollution through more pollution. Second, Reynold’s argument fails to consider the precautionary principle contained in the LRTAP protocols. Given the uncertainties around the effects of direct sulphur injection into the stratosphere, this principle is relevant to any assessment under the Convention, at least for the parties to the Protocols.

2.8 Law Creating Fora for Future Discussion and Regulation

79Different legal regimes have created governance mechanisms and forums for negotiations, which might be used to discuss and create regulation on the three proposed emergency technologies. These fora can notably be found in climate change law, the UNEA, the ENMOD Convention, and the ILC.

80As the primary instrument addressing measures against climate change, the UNFCCC might offer an appropriate forum to discuss the place of emergency technologies in combatting climate change (Craik and Burns 2016, 12). In particular, the global stocktaking process created by the Paris Agreement might be an opportunity to address this issue (Paris Agreement, art. 14. See also Craik and Burns 2016, 11-12). Moreover, the Subsidiary Body on Scientific and Technological Advice (SBSTA) might take up such discussions based on its mandate to provide advice on new technologies (Nicholson, Jinnah, and Gillespie 2018, 329. See UNFCCC, art. 9(2)(c); Paris Agreement, art. 18(1)). If emergency technologies are addressed in the UNFCCC framework, one might expect the stance of negotiators and secretariat staff to be rather sceptical given the regime’s strong commitment to mitigation and adaptation (Reynolds 2019b, 139). For instance, the SBSTA has, to date, focused on considering low-emissions technologies and technologies used for adaptation (Nicholson, Jinnah, and Gillespie 2018, 329). For artificial islands as an adaptation measure, the Cancun Adaptation Framework and the Warsaw International Mechanism for Loss and Damage are additional fora available for discussion beyond the global stocktake (Paris Agreement, arts. 7(7), 7(14) and 8(2)).

81The United Nations Environment Assembly (UNEA) provides another forum whose discussions might cover, in a general manner, all of the proposed technologies. In 2019, Switzerland proposed a draft resolution on geoengineering and its governance, which would have created an expert group to prepare an assessment of SRM and CDR technologies such as SAI and ocean fertilisation (Switzerland 2019). During the 2019 UNEA meeting, the proposal was discussed controversially. For instance, the appropriateness of the UNEA as a forum was questioned. Given the UNEAs’ mandate, discussions in the UNEA are naturally focused more strongly on the environmental risks of technologies than on their potential contribution to climate change policy (Honegger 2019). Some states thus favoured a more climate-focused and potentially more technology-friendly forum such as the UNFCCC or the IPCC, while others preferred an even more precautionary forum such as the CBD (Jinnah and Nicholson 2019, 877; McLaren and Corry 2021, 28). Moreover, the insistence of some, and opposition of others, to include a reference to the precautionary principle contributed significantly to the ultimate failure to reach a consensus on the resolution (Jinnah and Nicholson 2019, 877). Finally, controversies ensued when some states opposed including language underlining that geoengineering technologies are no alternative to mitigation (McLaren 2019). The combination of these controversies ultimately led Switzerland to withdraw its proposed resolution (Proceedings of UNEA 2019, para. 85; Paul et al. 2019, 13). Nonetheless, the UNEA might remain a relevant forum for discussing emergency technologies; Switzerland has promised to reopen the discussion at future meetings (Jinnah 2019).

82The Convention on the Prohibition of Military or Any Other Hostile Use of Environmental Modification Techniques (ENMOD Convention) has also been proposed as a venue for further discussion of SAI (McGee et al. 2020). Ocean fertilisation and the construction of artificial islands might potentially also be covered by the treaty regime, as they could equally be deemed environmental modification techniques. However, as the Convention does not cover peaceful use of environmental modification, it is mostly considered not to be currently applicable to technologies if they are not used in the context of an armed conflict (Bodle 2010, 312. See also ENMOD, art. III(1)). Nevertheless, some authors argue that states affected by side effects might be justified in considering any use of such technologies without their consent to be hostile (Lin 2013, 186).

83If one considers the use of SAI for addressing climate change to be peaceful in principle, the ENMOD Convention might still provide a forum for the exchange of scientific and technological information. The Convention explicitly obliges parties to exchange such information on peaceful environmental modification (ENMOD, art. III(2); McGee et al. 2020, 299-300). Furthermore, even if the ENMOD Convention has not yet been applied to any climate technologies, it has served as an inspiration for other regimes’ explicit regulation of climate modification technologies. For example, the ILC’s draft guideline on intentional large-scale modification of the atmosphere and the LP amendment on marine geoengineering have partly based their definitions on the ENMOD Convention’s definition of environmental modification techniques (Report of the ILC 2018, 181-182; Reynolds 2019a, 132). Nevertheless, in order to directly regulate new technologies and the conditions for their use, an amendment to the Convention would probably be necessary (McGee et al. 2020, 300-301). A challenge in leveraging the ENMOD Convention for the regulation of new technologies lies in the fact that it does not have a secretariat and that meetings of the parties only occur irregularly (McGee et al. 2020, 302-303; Reynolds 2019a, 132).

84Specifically for SAI, the CLRTAP might provide channels to address the technology and its side effects. Notably, mandatory consultations would enable parties that are affected or concerned about SAI’s side effects to request information from states experimenting with or using such technologies (CLRTAP, art. 5. See also Reynolds 2019a, 98). Parties to the Oslo or Gothenburg Protocol could also request the mandatory establishment of a conciliary commission if a dispute arises from a party’s use of SAI (Oslo Protocol, art. 9(5); Gothenburg Protocol, art. 11(5)). Additionally, parties or the secretariat might also refer concerns arising from SAI use by a party to the CLRTAP Implementation Committee, which could then issue recommendations to the Convention’s Executive Body (Decision 1997/2. See also Reynolds 2019a, 99).

85Finally, the ILC’s work on the protection of the atmosphere might give new momentum to discussions on SAI and ocean fertilisation, while its work on sea-level rise might spark legal development around artificial islands. Moreover, while not constituting a legal forum, the IPCC’s scientific consideration and discussion of different climate technologies in upcoming reports might significantly influence future discussion in legal for a (Jinnah and Nicholson 2019, 878). The IPCC might influence such discussion both through explicit consideration of the technologies and through its climate and mitigation models, which may make such technologies appear more or less necessary to achieve the objectives of climate change law (Jinnah and Nicholson 2019, 878; McLaren and Markusson 2020). Similarly, discussions in academia or civil society about climate technology governance might influence future binding regulation in legal fora. Such discussions have notably led to the so-called Oxford and Asilomar Principles on geoengineering governance, which consider that governance of such technologies needs to be in the public interest and underline public participation and transparency in research and regulation (Rayner et al. 2013; Asilomar Scientific Organizing Committee 2010). The UK parliament and government have endorsed the Oxford Principles (Rayner et al. 2013, 500). Nonetheless, the two sets of principles have not yet found serious consideration in multilateral discussions on climate technologies.


1 Parts of the discussion of artificial islands in this chapter are based on a paper entitled “We’re Moving!”: Relocating Sinking Small Island States to Artificial Islands?, which the author of the present paper submitted to Professor Anne Saab in fall 2019 for the seminar Climate Change and International Law at the Graduate Institute.

2 Du points out that the estimated yearly global sulphur injections required for effective SAI deployment are more than three times higher than the emissions ceiling of the European Union.

3 Currently, only six parties have accepted the amendment. See Overview of the Status of the London Convention and Protocol and Their Respective Amendments 2019; Acceptance by Germany of the 2013 Amendments 2020. See also the provision on entry into force of amendments: London Protocol, art. 21(3).

4 According to Ginzky and Frost 2014, 88, the qualifier ‘as far as practicable’ only applies to the obligation to prevent pollution, not the obligation to reduce pollution.

5 Submission of long-term climate strategies is mandated by Paris Agreement, art. 4(19).


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search