Climate Engineering rückt weltweit immer stärker in den Fokus von Wissenschaft und Politik. Es umfasst eine Bandbreite möglicher Techniken zu zwei wichtigen Aspekten: Sonnenstrahlung in den Griff zu bekommen und Kohlendioxid aus der Atmosphäre zu entfernen. Die zurzeit diskutierten Möglichkeiten haben enormes Potenzial, erheblich zur Treibhausgasreduktion beizutragen, sind allerdings mit Risiken und einer Reihe von Unwägbarkeiten verbunden.
Die "Europäische Transdisziplinäre Einschätzung des Climate Engineering" (EuTRACE), finanziert durch das 7. Rahmenforschungsprogramm, soll Climate Engineering aus europäischer Perspektive anwenden und mit den in Europa gesteckten Klimazielen verbinden sowie Chancen, Risiken und Auswirkungen untersuchen.
Climate change is a cross-cutting issue, involving broad aspects of science and society. Consequently, holistically assessing the potential benefits and perils of climate engineering requires a matching range of competencies across a variety of fields.
For this purpose, the project will be coordinated by the Institute for Advanced Sustainability Science (IASS), which was started in 2009 by founding director Klaus Töpfer, former executive director of UNEP, with three initial clusters focusing on broad issues relevant to global environmental change: society (led by Klaus Töpfer), energy (Carlo Rubbia, Nobel Laureate in Physics), and the atmosphere (Mark Lawrence, coordinator of EuTRACE). The IASS has gathered several top scientists in the field of climate engineering working at world-class institutions in Germany, the UK, Norway, France and Austria as partners to jointly address the natural science and social challenges of climate engineering. The following partners will work on EuTRACE:
• From Germany: The IASS, the Kiel Earth Institute (KEI), the KlimaCampus Hamburg (consisting of 18 research institutes including the Max Planck Institute for Meteorology), the Karlsruhe Institute of Technology (KIT) and adelphi; principal investigators include Hauke Schmidt (solar radiation management, coordinator of the EU FP7 IMPLICC project), Thomas Leisner (atmospheric science), Andreas Oschlies (carbon dioxide removal and oceanography), Gernot Klepper (economics of climate engineering), Alexander Proelß (international environmental law and governance), Jürgen Scheffran (risk analysis and security policy), Achim Maas (policy engagement and outreach), and Gregor Betz (philosophy and argument analysis).
• From the UK: The Tyndall Centre for Climate Change Research (headquartered at the University of East Anglia), the University of Exeter, Bristol University and the University of Edinburgh; principal investigators include Jim Haywood (atmospheric science, in particular aerosols), Matt Watson (atmospheric natural hazards), Tim Lenton (earth system modeling and co-evolution of life and environment), Stuart Haszeldine (carbon capture and storage), Tim Rayner (climate governance), Naomi Vaughan (atmospheric science, science/policy interface and dissemination), and Simon Shackley (innovation and technology studies).
• From Norway: the University of Oslo, the Norwegian Meteorological Institute and the Center for International Climate and Environmental Research (CICERO); principal investigators include Michael Schulz and Jon Egill Kristjansson (both atmospheric science), Asbjörn Aaheim (economics), and Anne Therese Gullberg (political science).
• From France: the Centre National de la Recherche Scientifique; the principal investigator will be Olivier Boucher (Earth System modelling), who is a coordinating author of the Intergovernmental Panel on Climate Change (IPCC) sections on climate engineering in the Fifth Assessment Report.
• From Austria: The University of Graz; the principal investigator Lukas Meyer (philosophy) will round out the consortium with expertise on the ethics of climate change and climate engineering.
These partners will make use of their global networks in the science and policy areas to achieve the widest possible dissemination and to subject the project findings to critical review by peers around the world.
ENV.2012.6.1-5 Explore opportunities, risks, feasibility and policy implications associated with key geo-engineering options - FP7-ENV-2012-one-stage Deliberate large-scale manipulation of the earth-climate system (geo-engineering) is increasingly explored as an additional potential strategy to counteract anthropogenic climate change. However, geo-engineering options i) suffer from limited understanding of the physical science basis; ii) include major uncertainties regarding effectiveness, impacts and feasibility; iii) lack comprehensive risk assessment.
The action should evaluate the main geo-engineering options in an inter-disciplinary manner, using the latest scientific data and information, in order to assess: i) whether, and if so how, they can effectively contribute to climate change mitigation; ii) their potential impact and associated risks on human and natural systems; iii) their feasibility including costs; iv) the associated governance and legal issues. The action should also identify key knowledge gaps and recommend future research needs. Social and policy implications that are likely to arise from the implementation of these options should also be explored. The action should take stock of the results of previous EU projects in the field. Funding scheme: Coordination and Support Action (supporting action) The requested EU contribution per project shall not exceed EUR 1000000 Up to one proposal can be selected.
Expected impact: Inform policy makers and the public about the main geo-engineering options in light of their effectiveness, risks, uncertainties, costs and governance implications. Better consensus on knowledge gaps and research needs (both at short and long term).
Final report of the FP7 CSA project EuTRACE
2.2 Albedo modification and related techniques
The Earth’s climate depends upon the balance between absorbed solar radiation and emitted terrestrial radiation (see Figure 1.2 for reference). Albedo modification refers to deliberate, large-scale changes of the Earth’s energy balance, with the aim of reducing global mean temperatures. The proposed methods are designed to increase the reflection of solar (shortwave) radiation from Earth. Suggestions for increasing the Earth’s reflectivity include: enhancing the reflectivity of the Earth’s surface; injecting particles into the atmosphere, either at high altitudes in the stratosphere to directly reflect sunlight or at low altitudes over the ocean to increase cloud reflectivity; and placing reflective mirrors in space.
Albedo modification techniques are distinct from mitigation and from most greenhouse gas removal techniques, in three key ways: their operational costs are potentially low; their effects are potentially rapid and large; their evaluation is better characterised as a risk-risk trade-off (Goes et al., 2011). In light of this distinction, various potential roles for albedo modification have been proposed: employing albedo modification on a large scale with the goal of reducing climate risks as much as possible, potentially substituting for some degree of mitigation (Teller et al., 2003; Carlin, 2007; Bickel and Lane, 2009); employing albedo modification as a “stopgap” measure to allow time for reducing emissions (Wigley, 2006); reserving albedo modification for use in a potential “climate emergency”, such as the large-scale release of methane from permafrost and ocean deposits (Blackstock et al.; 2009, see also Box 3.7).
Of course, it is also possible that albedo modification techniques will have no role in future responses to climate change, if it is decided not to employ any of them at all, given that they do not address the fundamental cause of global warming, namely emissions, and thus the increasing atmospheric concentrations of greenhouse gases (Matthews and Turner, 2009). Discussions on the potential role of albedo modification frequently focus on three key drawbacks. Firstly, since albedo modification impacts the climate in a manner that is physically different from the impact of greenhouse gases, it would not be possible to simply reverse the effects of global warming. Thus, whilst albedo modification may reduce some risks associated with climate change, it may in turn increase others. The way in which an albedo modification technique is deployed would affect the distribution of benefits and harms (Irvine et al., 2010; Ricke et al., 2010a; Mac Martin et al., 2013). Secondly, albedo modification carries the risk of a “termination shock”: if it were deployed for some decades at large scale and thereafter terminated, there would be a rapid warming globally, back towards the temperatures that the Earth would have already reached in the absence of a deployment of albedo modification (Matthews and Caldeira, 2007; Irvine et al., 2012). Such an event would likely be particularly damaging, given that there are indications that the impact of climate change on human populations and ecosystems depends strongly on not only the amount but also especially on the rate of climate change (Goes et al., 2011). Thirdly, albedo modification does not address the direct effects of CO2 on the environment, such as ocean acidification and impacts on terrestrial vegetation (Matthews and Caldeira, 2007).
Finally, beyond these physical risks, it is also important to note that the potential future role of albedo modification, if any, will also depend on how the initial scientific results are interpreted, framed, and com- municated, as well as on how the socio-technical context, into which discussions of climate engineering are emerging, shapes these techniques and their usage. This is discussed further in Chapter 3.
The Sections below (2.2.1 – 2.2.5) assess several of the key methods that are currently being discussed, which mostly involve increasing the planetary albedo, either at the Earth’s surface, or in the atmosphere via modifying low-level clouds or stratospheric aerosol particles. Using space mirrors for climate engineering is not discussed in detail, since the technological development, material and energetic requirements, and associated operational costs would at present be so prohibitive that this technique is not realistically being considered for implementation in the mid-term future, although it is used as a form of “thought experiment” for idealised climate model simulations, as discussed in Section 2.2.6. While the majority of studies have concentrated on albedo modification, a few other related techniques have been proposed that would alter the Earth’s energy balance by increasing the amount of terrestrial radiation emitted from the planet (see Section 2.2.5). Numerous additional methods beyond those discussed below have also been proposed. An overarching description of the research findings on the responses of the climate to the various methods is presented at the end of this section.
2.2.1 Stratospheric aerosol injection (SAI)
Description: Stratospheric aerosol injection involves increasing the amount of aerosol particles in the lower stratosphere (at altitudes above about 20 km) as a means to increase the reflection of sunlight beyond what is reflected by the naturally-occurring stratospheric aerosol layer (Niemeier et al., 2011; Rasch et al., 2008). Particles could either be injected directly or formed via injection of precursor gases such as sulphur dioxide (SO2), which are then converted into particles. SAI is currently the most discussed albedo modification technique. It was first proposed by Budyko (1974), but was not widely discussed until the idea was reiterated by Crutzen (2006); since then it has been heavily investigated, including numerous modelbased studies and first proposals and plans for field experiments, as well as studies on societal aspects such as perception, ethical concerns, and governance. Effectiveness: Analyses of the global temperature record subsequent to large volcanic eruptions that inject millions of tonnes of SO2 into the stratosphere leave little doubt that the introduction of aerosols into the stratosphere cools the climate (Robock, 2000). It has been suggested that the technical feasibility, effectiveness, affordability, and timeliness of stratospheric climate engineering techniques could make them a possible option for counteracting global warming (Robock et al., 2009), but many concerns have also been raised, including geographically inhomogeneous climate effects and potential side effects (Robock, 2008).
Candidate gases for injection into the stratosphere include SO2 or hydrogen sulphide (H2S) (Robock et
al., 2008; Shepherd et al., 2009; Rasch et al., 2008), which are oxidised to form small sulphuric acid aerosol particles (Robock et al., 2009). H2S has a lower molecular mass and is around twice as effective as SO2 in creating molecules of stratospheric sulphuric acid per kg of gas, but is highly toxic and thus may be problematic for transport to the stratosphere in large masses. Detailed aerosol microphysical modelling studies reveal a very complex picture of the interplay between injection rate, location, particle growth, and microphysical and optical properties; based on these, it appears likely that continuous injection of SO2 or H2S would lead to the formation of larger particles than observed subsequent to volcanic eruptions, if the total annual injection rates are comparable (Heckendorn et al., 2009; Timmreck et al., 2010; Niemeier et al., 2011). Direct injection of sulphuric acid, H2SO4, has also been proposed as a way to potentially gain more control over aerosol size, but given the higher molecular mass, a greater weight of material would then need to be lifted to the stratosphere (English et al., 2012; Pierce et al., 2010). Larger particles are less effective at scattering sunlight back to space and have a shorter atmospheric residence time because they sediment more rapidly, which means that one would expect diminishing returns in terms of the cooling impact for increased rates of injection (Niemeier and Timmreck, 2015). Such diminishing returns are demonstrated in Figure 2.4, which also shows that there are large differences between models in terms of the estimated injection rate needed to achieve a certain amount of radiative forcing. Particles of various chemical compositions have been proposed, including titanium dioxide (Pope et al., 2012), soot (Kravitz et al., 2012), and limestone (Ferraro et al., 2011), and nanoparticles that would photophoretically levitate, i.e., would be selflofting into the upper stratosphere (Keith, 2010). Customised particles could theoretically be designed to maximise the cooling impact while minimising side effects, for example on stratospheric ozone. Soot would dramatically heat the stratosphere, and would also decrease ozone levels and thus increase the UV flux at the Earth’s surface, and thus it has been suggested that it is not a suitable particle type or SAI (Kravitz et al., 2012).
Particles with a smaller radius and higher refractive index (for example titanium dioxide, TiO2) could be
around three times more effective than the same mass of sulphuric acid particles (Pope et al., 2012). However, coagulation of candidate solid particles in the stratosphere still needs to be comprehensively investigated. If the customised particles were to coagulate, particularly when mixed with the naturally occurring sulphate aerosol layer, then this would lead to much larger particles than those initially injected. This would in turn reduce their optical efficiency, increase absorption of terrestrial radiation, and increase their sedimentation rates. This may be of particular relevance should future large volcanic eruptions coat the candidate particles with a layer of sulphate.
Delivery mechanisms have received considerable attention. Estimates of the operational costs of various potential delivery mechanisms have been provided by Robock et al. (2009); Davidson et al. (2012); and McClellan et al. (2012). Those studies considered injection via: aircraft payloads, artillery shells, rockets, stratospheric balloons, and other possibilities. The main conclusion was that the most economically feasible injection mechanisms are likely to be injection via high-flying (higher than 15 km) aircraft, or by tethered balloons in the tropics. For aircraft delivery, installation cost estimates range from about $1 – 30 billion, plus annual maintenance costs in the range of $0.2 – 20 billion, depending on the injection strategy. Artillery shells and missiles were estimated to have maintenance costs in the range of 10 – 100 times more than those associated with aircraft delivery. Estimates for stratospheric balloon delivery are comparable to aircraft injections, in the range of $1 – 60 billion for installation costs with annual maintenance costs of $1 – 10 billion. The results of the initial studies vary widely, with some having nearly opposing trends. For example, Davidson et al. (2012) indicate maintenance costs for aircraft injections that are ten times those of balloon injections, whereas McClellan et al. (2012) suggest that balloon injections would instead be about ten times the maintenance costs of aircraft injections. These disparities are due to different underlying assumptions about the amount of material that is to be injected, with airplanes being more efficient at injecting smaller amounts and balloons becoming more cost-effective when larger amounts are injected (because they can continuously inject material, while airplanes can only inject a limited amount of material and would then have to pick up a new payload and refuel). There is, nevertheless, a broad consensus that dispersing megatons of particles into the stratosphere could be feasible and achievable at total operational cost that might be within the reach of individual nations, if the various practical constraints surrounding SAI delivery mechanisms were to be overcome (see Box 2.1).
The operational costs will depend on many factors, such as the altitude of deployment and the total amount of aerosol particle mass or precursor that is injected. The amount of sulphur that would need to be injected into the stratosphere to offset a warming corresponding to a doubling of atmospheric carbon concentration was estimated in Shepherd et al. (2009) to be between 1 and 5 Mt sulphur, while more recent estimates suggest that up to several tens of Mt could be required (Pierce et al., 2010; Niemeier et al., 2011). An important general consideration is that the forcing would not increase linearly with the injected mass, since greater injection rates will normally lead to larger particles that have shorter residence times and less favourable radiative properties, thus giving diminishing returns. As shown by Niemeier and Timmreck (2015), there may be an upper limit of the order of 10 – 15 W/m2 for the cooling that can be achieved by stratospheric aerosol particle injection.
Practical constraints surrounding SAI delivery mechanisms Tethered balloons are subject to constraints on the intrinsic mechanical properties of tethers, as well as to meteorological phenomena such as horizontal and vertical wind shear, clear air turbulence, cumulus convection, icing, and lightning, which would influence the possible locations of year-round operating bases (Davidson et al., 2012). Delivery from aircraft by increasing the sulphur content in aviation fuel and using the current distribution of civil aircraft flight paths would be ineffective, since the injection altitude is too low and most of the injection would be too far north to result in significant global cooling (Anton et al., 2012). Furthermore, under this scenario, stratospheric aerosol particles would be confined to the Northern Hemisphere, with impacts on the patterns of tropical rainfall (Haywood et al., 2013). Thus, if aircraft were used to deploy stratospheric aerosols, a new fleet of high-flying aircraft dedicated to the task
would likely be the preferable option (McClellan et al., 2012).
Other impacts: Stratospheric ozone is affected by stratospheric aerosols. The eruption of Mount Pinatubo reduced the total global amount of stratospheric ozone by approximately 2 % (Harris, 1997), since stratospheric particles offer heterogeneous surfaces for chemical reactions that deplete stratospheric ozone (Tilmes et al., 2008). However, this effect is small relative to the existing ozone reduction caused by chlorofluorocarbons (CFCs), and will also decrease proportionally as the amount of available chlorine decreases (as CFC concentrations decrease in the future). Candidate particles could theoretically be chemically engineered (Pope et al., 2012) by coating them to have less impact on stratospheric ozone than sulphuric acid particles. Scattering by the injected particles would serve to reduce the amounts of harmful UV radiation reaching the surface of the Earth, but this would only partially compensate the increases in UV radiation resulting from ozone loss (Tilmes et al., 2012). Reductions in ozone and the absorption of terrestrial radiation by the stratospheric aerosols could change the heating rates in the stratosphere, with consequent impacts on the dynamics of the stratosphere (Ferraro et al., 2014). Post-Pinatubo modelling shows that the absorption of terrestrial radiation by sulphate aerosols altered the stratospheric dynamics so that the aerosol was transported more strongly into the Southern Hemisphere rather than remaining primarily isolated in the Northern Hemisphere (Young et al., 1994).
In addition to stratospheric ozone, tropospheric cirrus clouds may also be impacted by stratospheric aerosols injections when the particles sediment from the stratosphere into the troposphere. Global GCMsimulations (Kuebbeler et al., 2012) suggest that stratospheric aerosol injections would reduce ice crystal number concentrations in cirrus clouds by between 5 and 50 %, which would lead to enhanced global cooling, highlighting a potentially important feedback mechanism. However, this effect is highly uncertain and not supported by observations, which have shown abnormally high ice crystal number concentrations in cirrus clouds due to aerosol particles that originated from the Mt. Pinatubo eruption sedimenting from the stratosphere into the upper troposphere (Sassen et al., 1995).
SAI may also influence vegetation growth by reducing solar radiation at the surface, which will reduce photosynthesis. This might be partly counteracted by the increased fraction of diffuse- to direct-radiation
(Vaughan and Lenton, 2011) which would result in higher photosynthesis rates (Mercado et al., 2009; Gu et al., 2002). The aforementioned ozone depletion might harm plants by letting more harmful UV radiation reach them, rather than being absorbed at high altitudes (Stapleton, 1992), although the aerosol itself will reduce levels of harmful UV radiation. Temperature changes on the regional scale will also influence vegetation, which also depends on how SAI is implemented: warming in polar regions might lead to enhanced vegetation growth, whereas in other regions it might have a different effect. Due to its effects on the global atmospheric circulation, SAI will change water availability. This may be especially important in water-limited regions, potentially increasing water availability and hence vegetation growth despite a globally weaker hydrological cycle. However, model simulations suggest that, due to elevated atmospheric CO2 concentrations, vegetation growth would increase significantly and that SAI would not change this by much (Bala et al., 2002; Naik et al., 2003; Jones et al., 2011).
Taken together, the various feedback mechanisms mentioned here make it challenging to anticipate the
distribution of stratospheric aerosols, their impacts on other components of the climate and Earth system (such as ozone and cirrus clouds), and hence the overall resultant environmental consequences of an injection of stratospheric aerosol particles or their precursors.
2.2.2 Marine cloud brightening (MCB)/ marine sky brightening (MSB)
Description: First proposed by Latham (1990), this method involves using water-soluble particles to seed low-level warm clouds over the oceans. These clouds tend to have a net cooling effect on the climate because of their strong reflection of solar radiation. The aim of this technique would be to increase their albedo, as well as to possibly also increase their lifetimes. This could be achieved by injecting suitable particles into the cloud updrafts, whereupon the particles act as cloud condensation nuclei, around which water vapour can condense. Increasing the number of cloud condensation nuclei increases the number of droplets in the clouds, meaning that the available water forms into more and smaller droplets than in non-seeded clouds. Since the ratio of surface area to volume increases for smaller droplets, the smaller droplets have a greater surface area and thus reflect more sunlight than fewer, larger droplets, as depicted in Figure 2.5. In addition to this impact on reflection by cloud droplets, the increased concentration of aerosol particles also directly causes an increase in the reflection of sunlight; however, the indirect effect via cloud droplets is usually larger than the direct effect via the much smaller aerosol particles themselves.
For seeding clouds, the type of aerosol particle most commonly suggested is naturally occurring sea salt. Since aerosol particle residence times are much shorter (days) in the lower troposphere near the Earth’s surface than in the stratosphere (order of 1 – 2 years), the particles would need to be continuously replenished and large masses would need to be injected in order to be effective. Recent model studies have shown that injection of sea salt could also be effective over cloud-free, low-latitude oceans, since the sea salt particles themselves also reflect solar radiation. The relative effectiveness of marine cloud brightening and this “clear-sky sea-salt effect” is currently not well known, but they appear to be of the same order of magnitude (Partanen et al., 2012; Jones and Haywood; 2012, Alterskjær et al., 2013). Throughout the report, the combined method is referred to as “marine sky brightening (MSB)”.
Effectiveness: Published estimates indicate that marine cloud brightening could potentially exert a radiative forcing of -1.7 to -5.1 Wm-2 (Jones et al., 2011; Partanen et al., 2012; Alterskjær et al., 2012; Latham et al.2008; Lenton and Vaughan, 2009; Rasch et al., 2009; Alterskjær and Kristjánsson, 2013). This high potential effectiveness is due to the extensive regions over which the technique could be applied: about 17.5 % (89.3 × 106 km2) of the Earth’s surface area is covered by marine stratiform clouds (Latham et al., 2008) and cloud-free areas over the subtropical oceans represent another 5 – 10 % of the Earth’s surface area. The method would be most effective when there are no or few higher-level clouds overlying the targeted low clouds and the air is unpolluted (Alterskjær et al., 2012; Bower et al., 2006). The effectiveness would be dependent on meteorological conditions such as horizontal and vertical wind speeds, as well as the formation of precipitation (Wang et al., 2011). The meteorological conditions could also influence the efficiency of the spraying process (Jenkins et al., 2013). The resulting size of the sea salt aerosol particles is crucial. For particles that are too large, cloud seeding might even lead to a reduction in cloud droplet number concentration, opposing the desired effect (Alterskjær and Kristjánsson, 2013; Pringle et al., 2012). Clouds are among the most complex and least understood components of the climate system, and there are large uncertainties associated with aerosol–cloud interactions (Korhonen et al., 2010; Stevens and Boucher, 2012) and aerosol–radiation interactions (Partanen et al., 2012).
The most commonly discussed delivery strategy involves wind-driven vessels that would pump sea spray into the overlying air, whereupon evaporation would form sea salt aerosols that would then be transported up to the base of overlying low-level clouds by below-cloud updrafts and turbulent motion in the marine boundary layer (Salter et al., 2008). Challenges associated with this method include coagulation of the injected sea salt particles into fewer and larger particles (that sediment more quickly), and reduced updraft speeds or formation of downdrafts due to cooling by the sea spray as it evaporates. One study showed that the former effect can reduce the cloud droplet number concentration by 46 % over emission regions, thereby reducing the effectiveness of the technique by almost a factor of 2 (Stuart et al., 2013), while the latter effect appears to be less significant (Jenkins and Forster, 2013). Despite these limitations, other delivery methods, for example using aircraft, are likely infeasible given the very large mass of cloud seeding particles required. The costs of research and development for unmanned floating vessels are estimated at about $100 million, whilst each ship is estimated to cost $1.5 – 3 million (Salter et al., 2008). Estimates of the number of vessels needed to achieve a 3.7 Wm-2 radiative cooling (equivalent to the forcing from a doubling of pre-industrial CO2) vary from ~1500 (Salter et al., 2008) to ~25000 (Alterskjær et al., 2013). The method is deemed to have a low technical feasibility, largely because of the difficulty of developing a reliable spray-generation technology that could efficiently produce particles of an appropriate size in sufficiently large quantities. More fundamentally, the level of understanding of the effects of such aerosol spraying is currently very low, as there are only a few numerical experiments available, of which the results are quantitatively quite divergent, for example due to differences in the treatment of sub-cloud updrafts (Jenkins et al., 2013).
Other Impacts: As for other forms of albedo modification, changes in the hydrological cycle are expected
for MSB (Robock et al., 2009; Jones et al., 2011; Rasch et al., 2009; Rasch, 2010; Jones et al., 2009; Baughman et al., 2012; Bala et al., 2010; Bala and Nag, 2011; Alterskjær et al., 2013). Particularly for implementations in the Pacific Ocean, there is the potential for changes in the El Niño Southern Oscillation (ENSO) due to the strong localised cooling that would need to be induced in one of the key target regions, off the coast of Peru (Baughman et al., 2012). MSB is expected to enhance precipitation over low-latitude land regions; this may enhance agricultural productivity in some regions but could also lead to increased flood risk (Bala and Nag, 2011; Alterskjær et al., 2013). The emitted sea salt could cause corrosive destruction of infrastructure and have detrimental effects on plants if local deposition rates were sufficiently high (Paludan- Müller et al., 2002; Muri et al., 2015).
2.2.3 Desert reflectivity modification
Description: Deserts are considered one of the most optimal areas for land-based reflectivity modification, due to low population density, a relatively stable surface, sparse vegetation, a large surface area (11.6 × 106 km2, 2.3 % of the Earth’s surface area, not including aeolian deserts), limited cloud cover, and low-latitude locations. A reflective material could be placed on desert surfaces, e.g., aluminium coated with polyethylene (plastic) (Gaskill, 2004), with the intention of increasing the albedo from the present value of 0.2 – 0.5 to about 0.8 (Tsvetsinskaya et al., 2002).
Effectiveness: Covering all deserts (i.e., 2.3% of the Earth’s surface) with material that would increase its albedo from 0.36 to 0.8 would give a maximum globally averaged radiative forcing of -1.9 to -2.1 Wm-2, assuming permanently clear skies in the desert regions (Lenton and Vaughan, 2009). The effectiveness of the method is likely to be reduced with time, as sand and debris are windblown onto the sheets, reducing their reflectivity, thus adding maintenance costs; for near-complete coverage, the logistics of access to the surfaces for such maintenance would present an additional challenge (although near-complete coverage would limit the amount of sand debris that might be mobilised by winds to coat the material). Installation costs have been estimated at ca. $0.3 per m2 of surface area (Shepherd et al., 2009), based on the simple assumption that costs would be comparable to painting human structures, with possibly comparable maintenance costs if the surfaces needed to be renewed or recovered on a regular basis. Under this assumption, the total cost for achieving -2 Wm-2 of forcing would thus be several trillion dollars, which is likely to be prohibitively costly for full-scale deployment. The technology to produce the plastic sheets already exists, but nevertheless the feasibility is low, due to the high installation costs and the challenging maintenance issues. Reversal of deployment would in principle be straightforward, since the sheets would be readily removable, although plastic waste would likely be a major issue, as polyethylene is not readily biodegradable. Other Impacts: The most significant impact expected from this technique would be a substantial perturbation of desert ecosystems, due to the physical coverage by the sheets and the reduced energetic input due to the increased albedo. No studies of these implications are known. Furthermore, the increased desert reflectivity is expected to lead to considerable regional climatic changes. Irvine et al. (2011) found large reductions in regional precipitation adjacent to deserts and a severe reduction in monsoon intensity due to increasing albedo in desert regions. Another potential impact is a substantial reduction in the nutrient supply to the Amazon, for which the Sahara is an important source (Koren et al., 2006; Swap et al., 1992). There has been very little research on this topic, and thus the present level of understanding is very low.
2.2.5 Cirrus cloud thinning
Description: In addition to the various schemes discussed above for increasing the planetary albedo, other related ideas for directly cooling the Earth’s surface have been proposed. Most discussed among these is the idea of cirrus cloud thinning. Cirrus clouds, like all other clouds, both reflect sunlight and absorb terrestrial radiation. However, cirrus clouds differ from other types of clouds in that, on average, absorption outweighs reflection, with the result that cirrus clouds have a net warming effect on the climate (Lee et al., 2009). It has been suggested that seeding the cirrus-forming regions in the upper troposphere with relatively few, highly effective ice nuclei could induce a fraction of the haze droplets in cirrus clouds to freeze by interacting with the ice nuclei, forming fewer and larger ice crystals (Mitchell and Finnegan, 2009). These ice crystals would quickly grow large, at the expense of smaller, supercooled water droplets, thus more rapidly sedimenting out of the clouds and thereby reducing the optical thickness and lifetime of the seeded cirrus clouds (see Figure 2.6). Seeding material would need to be added regularly, because it too would fall out together with the large ice crystals. Reducing the optical thickness and lifetime of cirrus clouds would increase outgoing terrestrial radiation, causing a cooling effect. Bismuth tri-iodide, BiI3, has been suggested as a seeding material, as it is relatively cheap and non-toxic (Mitchell and Finnegan, 2009). Sea salt has also been suggested as another potential seeding candidate (Wise et al., 2012). The seeding aerosols can have an atmospheric residence time of up to 1 – 2 weeks, depending on their size and thus their sedimentation velocities. Commercial airliners and unmanned drone aircraft have been suggested as potential delivery mechanisms (Mitchell et al., 2011). The seeding substances could be dissolved into the jet fuel, or a flammable solution could be injected into the jet engine exhaust. Given that global cirrus cloud coverage
is 25 – 33 % (Wylie et al., 2005) (128 – 168 × 106 km2), large areas of the Earth would, in principle, be
susceptible to modification.
Effectiveness: A net cloud forcing of up to -2.7 Wm-2 has been found in model studies (Storelvmo et al., 2013; Muri et al., 2014). The method would be most effective at high altitudes (~10 km), in air with low background aerosol particle concentrations, and at night-time. The method would also be most effective outside of the tropics, since ice crystals in tropical cirrus clouds (typically anvil clouds) are predominantly formed in strong updrafts in convective clouds, making seeding a challenge. Combined with recent findings of heterogeneous nucleation in tropical anvil cirrus (Cziczo et al., 2013), this effectively rules out tropical cirrus for seeding (Storelvmo and Herger, 2014). Very little is currently known about the feasibility or operational costs of this approach; there are very few theoretical studies, and to the extent of our knowledge, no experimental field studies or implementations have been attempted to date. The maintenance cost of procuring the seeding material, BiI3, is likely negligible given the relatively small mass required (140 tonnes, Marshall, 2013), although there are other, likely more expensive, components of the total operational costs, for example aircraft deployment in susceptible regions, for which no estimates are yet available.
Other Impacts: In the context of cirrus cloud thinning, the cloud–aerosol–climate interactions are not well
understood. Factors that control the heterogeneous freezing process are uncertain, as ice growth kinetics are not well documented. “Over-seeding” might lead to warming, as opposed to the desired cooling (Storelvmo et al., 2013). Vertical velocities are important for activation of ice nuclei, but current estimates are uncertain due to lack of observations. Heterogeneous freezing may already be common in cirrus (Cziczo et al., 2013), which would render the method less effective than expected. There could be a number of climatological side effects. However, as cirrus cloud-thinning targets terrestrial radiation — effectively directly countering the greenhouse effect by reducing the amount of terrestrial radiation that is re-radiated from the atmosphere towards the surface, albeit not with the same geographical distribution of radiative forcing by greenhouse gases — it may nevertheless reduce the degree of atmospheric circulation changes and regional changes to the hydrological cycle that would be expected with most of the albedo modification schemes (Muri et al., 2014). With few numerical experiments available, the current level of understanding is very low.
2.2.6 Results from idealised modelling studies
Prior to potential future field tests or implementation, an initial understanding of the climate response to
modifying the planetary albedo can be gained from both highly idealised studies and more realistic scenarios of deployment. Idealised studies have proven very useful as they allow individual aspects of the response to be clearly identified. However, these idealised experiments are inevitably oversimplifications and omit many of the subtleties that would be involved in the deployment of an albedo modification scheme. Realistic scenarios of deployment, in contrast, are important for discussions with stakeholders about the topic; however, since many factors generally influence the climate simultaneously in such simulations, it can be difficult to isolate the causes of various responses.
In many idealised numerical model studies to date, a reduction in the incoming solar radiation (solar irradiance) has been used as a simple proxy for the various forms of albedo modification. While it might be possible to achieve such a uniform reduction by placing an array of mirrors in space, this technology is not considered a realistic option in the foreseeable future. However, since such simulations involving a uniform reduction in incoming solar radiation are straightforward to set up and compare with each other, this approach has been favoured to enable many climate modelling centres to conduct the same simulations, providing more confidence in the results. This has been done under the auspices of GeoMIP (see Box 2.2), providing a valuable background knowledge base for future, more realistic and more complicated experiments that are being initiated in the next phase of GeoMIP.
3. Emerging societal issues
Beyond the technical challenges of understanding and possibly exerting some degree of control over the impacts of climate engineering on the Earth system as described in the previous chapter, techniques that have been proposed for removing greenhouse gases or for modifying the planetary albedo or cirrus clouds all raise complex questions in the social, ethical, legal, and political domains (Shepherd et al., 2009). This chapter assesses several of these societal issues, which to a considerable degree shape the debate around different climate engineering techniques. Even the mere ideas of greenhouse gas removal or albedo modification raise important questions, for example about the possible influence that considering such techniques may have on efforts toward mitigation and adaptation (3.1.1), or about the responsibility that humans have toward the environment (3.1.2). This chapter also considers public awareness of the different techniques, and how this has developed in recent years (3.1.3), drawing on a preliminary analysis of four field experiments or trials to assess potential avenues for future research and attention. A second part of the chapter assesses aspects of potential climate engineering deployment, such as political conflicts that may ensue (3.2.1), along with economic considerations (3.2.2.). As will be shown, this also raises normative questions of fairness and justice, based on the distribution of benefits and costs (3.2.3) and of compensation for harm (3.2.4). The different issues discussed in the first two parts make decisions on interventions in the climate system, as well as about research on such interventions, extremely challenging. Building on this, the assessment presented in Chapter 6 addresses some of the difficulties for decision making and draws on various principles to examine the possible directions that such decisions may take. Within the different subsections of the chapter, the issues are presented first in a general way and are then applied to the three techniques detailed in the previous chapters (BECCS, OIF, and SAI).
3.1 Perception of potential effects of research and deployment
There are several issues related to the manner in which the discourse around climate engineering is perceived by policy makers and the broader public, including possible responses to the perception — justified or not — of a near-future “solution” to climate change. This section focuses particularly on the responses to research on climate engineering and on various aspects of carrying out the discourse, e.g., participation and consultation.
3.1.1 Moral Hazard
A prominent concern around climate engineering has been the fear that discussing, researching, and developing climate engineering techniques may have negative effects on efforts to reduce emissions. In general, such concerns have been summarised under the term “moral hazard” (Keith, 2000), which originated in insurance theory (Arrow, 1963). These concerns prevented many researchers from engaging with the topic until an editorial by Paul Crutzen in the journal Climatic Change (Crutzen, 2006) served to break the “taboo” (Lawrence, 2006) perceived by many in the atmospheric research community. It has been suggested that the moral hazard effect may occur via several mechanisms, which may also interact. These include: increasing risk-prone behaviour; diverting attention, efforts, and incentives from the challenge of decreasing greenhouse gas emissions; encouraging political stagnation; supporting a cost/benefit-based delay of emission reductions; and worsening existing coordination problems in climate policy (see, for example, Hale, 2012, Lin, 2013, Preston, 2013).
Three background assumptions are often associated with the moral hazard argument. First among these is the danger that climate engineering techniques could be misused to protect the vested interests of involved actors. In particular, this involves the possibility that developing and applying climate engineering techniques could be used to strengthen the position of those who oppose emission reductions, especially by those that profit from fossil-fuel-intensive production processes and fossil fuel extraction (Jamieson, 1996, Virgoe, 2009: 105, Ott, 2012). Secondly, many have associated the array of possible responses to climate change with hopes for more fundamental changes to the current systems of production and consumption. Proposals for climate engineering, however, generally focus on modifying the physical environmental aspects of global warming rather than on the underlying societal causes. This would potentially decouple other societal issues such as consumerist lifestyles, mass agricultural practices, unsustainable processes of energy and consumer goods production, tropical deforestation, and population trends from the debate on global warming (Corner and Pidgeon, 2010; Schäfer et al., 2014). Finally, such concerns about maintaining the status quo can be linked to an underlying concern about the use of technological fixes, which are seen as being based on a deep-rooted habit of solving problems with technology by changing the external circumstances (for example, applying more technology) rather than by changing behavioural patterns (Borgmann, 2012). Even though such fixes are used in many areas, their moral status is generally considered ambiguous and the techno-fix framing is often used negatively, to connote an inadequate and morally problematic solution to an underlying problem (Preston, 2013).
Views sceptical of the moral hazard argument have also been set forth. Firstly, the ethical implications of the suggested relationship between mitigation and climate engineering techniques are unclear. As Preston (2013) points out, warnings of a moral hazard are ambiguous and cannot provide clear guidance because it is unclear what action should follow from such warnings. Secondly, behavioural change due to reductions in the risks one faces can be seen as a rational response to the emergence of a new situation. Often, the negative evaluation of such a behavioural change is linked to the specific characteristics of the behaviour for which insurance is sought, rather than the measure taken to reduce the risk (Hale, 2009). Thus, safety technologies that create or could create an increase in risk-prone behaviour are not always abandoned, at least not as long as the benefits are believed to outweigh the possible costs (Bunzl, 2009). What seems to make individual climate engineering techniques problematic is the assumption that they will contribute to an underestimation or intentional downplaying of the risks and uncertainties associated with emitting greenhouse gases, and thus to the transfer of those risks to others, especially future generations. Thirdly, very little empirical evidence is currently available in this area, and claims about a moral hazard may only be testable, if at all, if the situation ever actually develops in a substantial and observable form (Lin, 2012, Preston, 2013). Fourthly, it is unclear whether mitigation efforts would be more successful in the absence of discussions on climate engineering, as many different factors contribute to the political inertia against reducing emissions (Davies, 2011). Finally, some argue that the prospect of specific climate engineering techniques (especially SAI) could have the reverse effect, as the mere thought that they might be deployed might be perceived by some as being so threatening that they strengthen the support for mitigation (Davies, 2010, Davies, 2011, Moreno- Cruz and Smulders, 2010). Millard-Ball (2012) argues that it may be rational for other countries to respond to such a threat by reducing emissions to the level where a country that is threatening to deploy a technique such as SAI no longer perceives a necessity to do so.
The moral hazard arguments have been applied especially to SAI, which has been referred to as a fast and cheap “technological fix” (Corner and Pidgeon, 2010, ETC Group, 2010). Due to incomplete knowledge about potential side effects, high risks, and pervasive uncertainties, there is very little support in the literature for SAI as a replacement or substitute for mitigation (Rickels et al., 201; Shepherd, 2009).
Greenhouse gas removal techniques such as OIF and BECCS might also create a moral hazard, which
would have both similarities and differences to the moral hazard potentially created by SAI. Here the focus is not on the possibility of rapidly counteracting specific effects of climate change, but rather on the potential for developing techniques, or for creating the perception that it will be possible at some point to develop such techniques, that might help address the physical causes of global warming at a later point in time. Given the prominent role that greenhouse gas removal techniques are given in the mitigation pathways underlying RCP 2.6 (and to some extent RCP 4.5), this concern seems particularly noteworthy (see Section 2.1.2). Even though the effectiveness of many techniques that aim to remove greenhouse gases from the atmosphere seems questionable in light of the scale of current emissions and the concentrations of atmospheric greenhouse gases that might occur in the future, the mere perception that greenhouse gases might be removed from the atmosphere at a large scale may still lead to reduced efforts towards mitigation. Techniques for removing greenhouse gases may therefore also exacerbate carbon-based path dependency in the near term (Unruh and Carrillo-Hermosilla, 2006).
3.1.2 Environmental responsibility
The potential use of different climate engineering techniques that have large-scale influences on the climate system has been ascribed various negative character traits, including hubris, arrogance, and recklessness (Kiehl, 2006, Hamilton, 2013). Some of these arguments are concerned with the potential negative influence of specific climate engineering techniques on humanity’s relationship to nature (Ralston, 2009). From this perspective, such techniques may not only have adverse effects on the environment, but could also exacerbate a perceived lack of environmental responsibility (Buck, 2012). In this sense, the intentional manipulation of the climate has been described as a further instance of humans’ unwillingness to “live with nature” (Jamieson, 1996) and the “crossing of a new threshold on the spectrum of environmental recklessness” (Gardiner, 2011). This also hints at the potential hubris toward human capabilities in which domination or control over natural processes is sought (Ralston, 2009; Joronen et al., 2011). Concerns have been raised (Matthews and Turner, 2009) that, due to the potential for human error and unintended consequences of such interventions, the claim of controllability
created by overconfidence and “appraiser’s optimism” may turn out to be an illusion (Amelung and Funke, 2013). Nevertheless, the use of terms such as hubris may be misleading. Individual climate engineering techniques differ greatly with regard to their novelty, scalability, and expected environmental impacts (Preston, 2013, Heyward, 2013), so at a minimum an explicit differentiation between individual techniques is necessary. It is also debatable whether direct interventions in the climate system need to be understood as transcending a threshold in our relationship to the Earth (Ridgwell et al., 2012). Some argue that humanity is already engaged in a largescale experiment with the climate through the use of fossil fuels (Davies, 2011). Here, the key point concernsan ethical distinction between intentional and unintentional interventions in natural systems and processes, and what this means for our responsibility. This is a discussion that has just started to emerge (Jamieson, 1996, Tuana, 2013).
3.1.3 Public awareness and perception
The literature on public awareness and public acceptance of climate engineering techniques is recent and diverse. Since awareness is typically assumed to be a precondition for the formation of beliefs and attitudes toward a novel technology, social science research has investigated both public awareness and acceptance of different climate engineering techniques and proposals. Empirical studies have used a variety of methods, including questionnaires and focus groups, with research designs including correlational analyses of surveys (e.g., Bellamy and Hulme, 2011), a quasi-experimental study (Kahan et al., 2012), and deliberative focus groups (e.g., Macnaghten and Szerszynski, 2013). Given that awareness of climate engineering is particularly low (results of a UK-based study (Pidgeon et al., 2012) show that 75 % of the respondents in the national sample had either “not heard of” the term climate engineering or knew “almost nothing about it”), some research has employed methods of public dialogue as a means to both raise awareness and to solicit attitudes from participants (Bellamy et al., 2013; Macnaghten and Szerszynski, 2013).
The literature has had varied foci, ranging from climate engineering in general to specific techniques or
groups of techniques, particularly planetary albedo modification (e.g., Mercer et al., 2011). Typically, techniques for greenhouse gas removal are not discussed in isolation, but within the context of contributions on climate engineering in general and as part of the literatureon CCS, including BECCS. Additionally, some studies have primarily focused on perceptions of climate change, within which climate engineering has been incorporated as one element (e.g., Bostrom et al., 2012); or on the perceptions of possible responses to climate change — one being climate engineering (Poumadère et al., 2011).
Studies of public awareness suggest that wording is important. The use of the term “climate engineering” was associated with higher levels of public awareness than use of the term “geoengineering” (Mercer et al., 2011). In this context, it is counterproductive to think of individuals as being similar to “empty vessels” that need to be filled with scientific/factual information about climate engineering. The findings of qualitative and deliberative research (e.g., Pidgeon et al., 2012) indicate that members of the public associate climate engineering with diverse, complex ideas, including “messing with nature”, science-fiction, “Star Wars” and environmental dystopia.
Two studies that applied deliberative methods suggest a position of “qualified” or conditional acceptance, in which support for research into climate engineering may not correlate with support for actual deployment (Parkhill and Pidgeon, 2011, Macnaghten and Szerszynski, 2013; Pidgeon et al., 2013). Mercer et al. (2011) found some acceptance of research on albedo modification techniques in a web-based survey in theUK, US, and Canada. Support was found to decrease when respondents were asked about using such techniques immediately, or to stop a climate emergency, while respondents disagreed when asked whether such interventions should ever be undertaken. Pidgeon et al. (2012) found greater support in the UK for techniques to remove greenhouse gases than for albedo modification, which they suggest may be linked to concerns about “interference with nature” (see Section 3.1.2), as well as issues of reversibility and
A study by Macnaghten and Szerszynski (2013) found public concern regarding the potential for international conflict following unilateral actions by nationstates, together with scepticism concerning the ability of national governments and international institutions to effectively govern techniques such as SAI in light of the slow progress on coordinated climate change mitigation. Public uncertainty regarding the value of field trials is a consistent finding, revealing doubts about avoiding unintended consequences to weather systems, as well as concerns about being fully able to predict the impacts of large-scale deployment following a relatively small-scale field trial (Parkhill and Pidgeon, 2011). Research on small-scale field trials suggest similarities with previous research on the “NIMBY” (Not In My Back Yard) concern of local communities, including issues of local governance regarding site selection and public consultation (Parkhill and Pidgeon, 2011; Pidgeon et al., 2013). Informing public audiences about past geopolitical attempts to shape weather and climate, and allowing deliberative discussions of the implications of this, appears to have the potential to lead to a hardening of participants’ attitudes towards consenting to any research on albedo modification (Macnaghten and Szerszynski, 2013).
This demonstrates the significance of “framing effects”: public acceptance of climate engineering will be influenced by how it is presented and by the issues or technologies with which it is associated. Although these studies begin to indicate the complex array of worldviews, values and beliefs that are likely to influence public perception of climate engineering, conclusions based on their findings are necessarily tentative. Previous studies employed differing methodologies and independent variables; consequently, replication is required to corroborate their findings and to determine the dependence on regional and cultural contexts. Low levels of public awareness would suggest that attitudes may not yet have formed for many people and even if they have, those opinions are likely to be weak and unstable. For this reason, some have suggested that the use of qualitative research and deliberative methods of public engagement may be the more suitable methodological approaches to adopt (Corner et al., 2012).
By comparison, a more substantive literature exists on public acceptance of CCS, including several studies focusing on the case of BECCS (see section 3.1.4). Some of these studies have compared attitudes to CCS with those surrounding renewable energy (Oltra et al., 2010; Upham and Roberts, 2011a; Scheffran and Cannaday, 2013), indicating that public acceptance of the latter is higher than for the former. They also illustrate doubts about whether CCS could contribute to solving the climate change problem. Palmgren et al. (2004) report mixed results on attitudes to CCS, correlating with views on anthropogenic climate change. Providing more information resulted in stronger opposition to CCS, especially against storage in the ocean. They conclude that public acceptance of CCS would require prior acceptance of the climate change problem. Huijts et al. (2007) find a slightly positive attitude toward carbon storage in general, whereas Oltra et al. (2010) find that the dominant public view of CCS is sceptical due to perceived risks. Terwel and Daamen (2012) point out that NIMBY effects do not necessarily dominate initial reactions to CCS. Several contributions on CCS emphasise the importance of trust in government and in the actors involved (Upham and Roberts, 2011b; Itaoka et al., 2012; Terwel and Daamen, 2012).
The effect of information upon public acceptance is uncertain (Fischedick et al., 2009; Itaoka et al., 2012). Huijts et al. (2007) find that information is not always asked for and does not necessarily increase acceptance — people that strongly object to technologies are often highly informed about them. As mentioned above, Palmgren et al. (2004) found that information provision increased resistance to CCS. Beyond the quantity of information provided, research suggests the importance of information qualities. When attitudes are weak or unstable, framing effects can play a significant role in shaping attitudes subsequently elicited. This poses a challenge for deliberative public engagement, since the a priori choices made by the research team regarding what information to provide to participants about climate engineering are likely to have a strong effect upon the results. Given this, increased use of experimental designs in future studies would be useful to test the impacts on public acceptance of providing specific forms of knowledge on climate engineering.
3.1.4 Participation and consultation: questions from example cases
This section examines four example cases (Box 3.2- 3.5) that were selected because of their potential relevance to questions that arise in the context of discussions of the three main techniques examined in this assessment: one field experiment examining OIF; two projects aimed at developing prototype implementations of BECCS; one project that included a planned field test of a delivery technology for albedo modification by SAI.
The exploratory assessment of these example cases brings to the fore questions about the role of risk assessment, the impact of private sector involvement on public perception, and the role of trust and public participation, as well as governance in climate engineering field experimentation.
The following sections distil questions that arise from these exploratory assessments of the example cases. While the exploratory assessments described above do not allow for making generalisable claims, their value lies in pointing to areas that deserve further attention and inquiry. For all areas described below, further research would be necessary to better understand the social dynamics involved.
What is the role of risk assessment in designing climate engineering field experiments?
Clearly distinguishing between what counts as experiment, test bed, demonstration, research, development, and deployment proves difficult. The experiences of this issue in the example cases indicate that openness and transparency about the intent behind an experiment, its scale, and possible routes to scaling up experiments to the level of demonstration or implementation play important roles in shaping public perception of various types of studies. Given the technical uncertainties and complexities in the example cases, and the breadth of stakeholder standpoints and motives, it is unlikely that all stakeholders involved would ever be fully satisfied with any particular risk assessment. This suggests it might be useful to develop principles for guiding decisions on whether or not to take forward a climate engineering experiment. Although different methodologies were used to frame, assess, and mitigate potential risks and impacts across the example cases (risk impact assessment and legal analysis in LOHAFEX; Underground Injection Control permit application for Decatur project, Illinois; use of the responsible innovation framework in SPICE), in all cases, novel assessment frameworks or modifications of existing frameworks were deemed necessary by those responsible for the projects.
What is the role of private sector interests in shaping public perceptions of climate engineering field experiments?
Personal and private-sector interests in a project, and how such interests are portrayed and perceived, may influence whether and to what extent a particular project can be realised. The existence of private sector or personal interests, whether as intellectual property rights on the part of individuals or the commercial interests of private companies, can become a source of conflict in the context of growing ethical and political debates around the commercialisation of and motivations for climate engineering research and technology development. The example cases suggest that, in order to prevent or resolve such conflicts, transparency and openness on intellectual property, and commercial or other vested interests may be beneficial. Entrenched views and scepticism concerning the will of private sector actors to work for the collective good may undermine consensus-building. Based on the example cases described here, it appears that building trust in project developers, project managers, and mediating institutions can prove fundamental to the acceptance of climate engineering research projects.
What is the role of trust and public participation in shaping public perceptions of climate engineering field experiments?
Public participation and engagement can be crucial to a project’s success. Local communities intensively opposed one of the field tests described above (Greenville) and were supportive in two other cases (Decatur and SPICE), whereas LOHAFEX was more remote from local communities. In the LOHAFEX and SPICE cases, it was predominantly international environmental NGOs and the media that played a pivotal role by drawing public attention to the activity, and in the case of LOHAFEX also questioning its legality. The example cases seem to suggest that early and ongoing public participation and engagement is an important consideration for experiments in this sensitive area of research, but at the same time does not and engagement requires relationships of trust between the different actors and stakeholders involved, which in turn often requires a history of institutions working together effectively.
What is the role of governance for climate engineering field experiments?
The application of some form of governance did not guarantee the success of the projects (two of the four example projects were not completed). Lack of clarity regarding the applicability of an international resolution (the UN CBD) led to controversy and to the adoption of new assessment procedures by the relevant scientific and research funding communities. Further detailed consideration of such regulatory
aspects is provided in Chapter 4.
3.2 Societal issues around potential deployment
While the preceding section has focused largely on climate engineering research, this section will consider issues surrounding the potential deployment of climate engineering techniques. Figures 3.1 and 3.2 illustrate that the deployment of climate engineering techniques such as SAI and BECCS may result in various potential impacts (black arrows) on physical, ecological, and social systems, possibly involving complex feedback processes (red arrows).
Figure 3.1 shows primary and secondary effects of climate engineering through SAI. According to model
computations, SAI is capable of reducing the global mean temperature, but would also reduce the global intensity of the hydrological cycle and change regional weather patterns (Kravitz et al., 2013a; Tilmes et al., 2013). Although this could potentially reduce overall climate risk at a global scale, it would change the distribution of regional and local risks. Beyond the intended effects on climate, SAI would also change stratospheric temperature and chemistry, for instance influencing the ozone layer (Heckendorn et al., 2009; Tilmes et al., 2009), and would also affect the occurrence of acid rain, although it is unclear whether this impact would be significant in comparison to that attributed to surface-level pollution sources (Kravitz et al., 2009). In addition to side effects on natural systems, SAI has potential public health impacts, for example by decreasing the thickness of the ozone layer, in turn increasing UV radiation at the Earth’s surface. It has also been suggested that the potential for conflict and social inequality could increase, for example through reductions in agricultural productivity and forestry or through dual-use problems that might arise from potential military applications, aggression and power plays associated with SAI techniques (Bodansky, 1996, Robock, 2008).
Figure 3.1: Schematic overview of possible consequences of the deployment of SAI. The legend on the bottom shows the colourcoded argumentation spheres: environmental (olive); scientific (light blue); economic (yellow); political (orange); social (brown); individual (violet). The boxes in the figure show various possible consequences of SAI deployment, with the colour of each box corresponding to the argumentation sphere that is most relevant to the consequence, and the colour of the line around the respective box corresponding to the second most relevant argumentation sphere. Grey arrows indicate plausible consequences; red arrows indicate feedbacks. The following sections of this chapter focus on three of the major argumentation spheres: economic, social, and political.
Source: Jasmin S. A. Link and Jürgen Scheffran, University of Hamburg.
Figure 3.2 shows primary and secondary effects of BECCS. The implementation of BECCS would compete with land use for food production, and may therefore contribute to social inequality (Lovett et al., 2009). Land use change also generally influences ecosystems, biodiversity and soil structure, for example if non-native species are favoured to maximise biomass growth (Chapin III et al., 2000; Sala et al., 2000; Dupouey et al., 2002). BECCS could even influence the regional weather and climate if applied on a large scale, for instance if deliberate deforestation were to increase the albedo (Bonan, 2008; Peng et al., 2014). Deforestation can also lead to local increases in particulate matter, with implications for human health (Beckett et al., 1998), as well as for regional clouds and climate. Another effect might be accumulated health risks or even sudden deaths, if there were ever to be a substantial CO2 leakage from underground storage sites (Solomon et al., 2007, 9ff.). Finally, the deployment of BECCS can also trigger local and national responses, for example due to the NIMBY effect (see Section 3.1.3), but also due to land use conflicts, increased food prices, or demands for subsidies for crop production.
Figure 3.2: Schematic overview of possible consequences of the deployment of BECCS. The legend on the bottom shows the colour-coded argumentation spheres: environmental (olive); scientific (light blue); economic (yellow); political (orange); social (brown); individual (violet). The boxes in the figure show various possible consequences of BECCS deployment, with the colour of each box corresponding to the argumentation sphere that is most relevant to the consequence, and the colour of the line around the respective box corresponding to the second most relevant argumentation sphere. Grey arrows indicate plausible consequences; red arrows indicate feedbacks. The following sections of this chapter focus on three of the major argumentation spheres: economic, social, and political.
Source: Jasmin S. A. Link and Jürgen Scheffran, University of Hamburg.
3.2.1 Political dimensions of deployment
Since each technique is situated in a specific sociopolitical context, it is important to reflect on how this
context might change through the emergence of that technique. Different political consequences would
emerge along the lifecycles (concept development, research, deployment, and various possible side effects) of the various proposed techniques, if they were to be pursued. Critical issues include the use of resources during the deployment process, the direct impacts upon the environments in which the techniques might be implemented, as well as unexpected consequences of the techniques on nature and society (e.g., Caldeira, 2012; Honegger et al., 2012; Klepper, 2012; Lin, 2012; McLaren, 2012; NOAA, 2012; Mooney et al., 2012; Bellamy et al., 2012).
To date, there has been no integrated analysis of the linkages between climate change, the different climate engineering techniques, and their combined effects on human security, conflict risks, and societal stability. Nonetheless, it has been argued that various conflict types may emerge throughout the lifecycle of climate engineering activities (Maas and Scheffran, 2012; Scheffran and Cannaday, 2013; Brzoska et al., 2012; Link et al., 2013). The following discussion distinguishes between five conflict types: competition over scarce resources; resistance against impacts and risks; conflicts over distribution of benefits, cost and risks; complex multi-level security dilemmas and conflict constellations; power games over climate control.
1) Competition over scarce resources: While many albedo modification techniques could likely be implemented with comparatively limited resources, most greenhouse gas removal techniques, such as BECCS or enhanced weathering, demand massive resource inputs to have a globally meaningful impact (see Section 2.1). Large-scale deployment of most techniques for the removal of greenhouse gases would need extensive infrastructures, thereby requiring the extraction and conversion of major resources (energy, raw materials, water, and land) that have an impact on natural and social systems in the affected regions. Thus, competition over physical resources; financial resources like investments; and immaterial resources such as human, social, and political capital could increase, affecting the availability of resources for mitigation and adaptation.
2) Resistance against impacts and risks: Anticipation of foreseeable or suspected impacts; detrimental side effects; and externalities such as pollution, modified rainfall patterns from SAI, or changes in ecosystems, vegetation, and crop yields; as well as principled opposition, might provoke resistance on local, national, and international scales. Furthermore, different techniques for the removal of greenhouse gases have specific local impacts (for example on water, biodiversity, forests, agriculture, or cities) that might encounter resistance from those who are affected and have inadequate coping mechanisms. Moreover, since technical, economic, and political limitations mean that some techniques are feasible only in certain regions, there may be particularly enhanced pressure on the resources and communities in these specific regions. This poses challenges for public acceptance and local coping mechanisms comparable to those associated with other forms of environmental modification, such as large-scale forest clearance for agricultural use, damming rivers, and the creation of artificial lakes (Conca, 2010, Balint, 2011).
3) Conflicts over distribution of benefits, costs, and risks: With its high leverage potential for shortterm effects on the global climate system, as well as potentially unpredictable variations in regional
impacts that might be adverse to local interests, SAI could provide ground for various conflicts. Given that, as discussed in Sections 2.2.1 and 2.2.6, temperatures would be decreased by different overall amounts depending on the amount of material that is injected, Finally, the deployment of BECCS can also trigger local and national responses, for example due to the NIMBY effect (see Section 3.1.3), but also due to land use conflicts, increased food prices, or demands for subsidies for crop production.
3.2.1 Political dimensions of deployment
Since each technique is situated in a specific sociopolitical context, it is important to reflect on how this context might change through the emergence of that technique. Different political consequences would emerge along the lifecycles (concept development, research, deployment, and various possible side effects) of the various proposed techniques, if they were to be pursued. Critical issues include the use of resources during the deployment process, the direct impacts upon the environments in which the techniques might be implemented, as well as unexpected consequences of the techniques on nature and society (e.g., Caldeira, 2012; Honegger et al., 2012; Klepper, 2012; Lin, 2012; McLaren, 2012; NOAA, 2012; Mooney et al., 2012; Bellamy et al., 2012).
To date, there has been no integrated analysis of the linkages between climate change, the different climate engineering techniques, and their combined effects on human security, conflict risks, and societal stability.
Nonetheless, it has been argued that various conflict types may emerge throughout the lifecycle of climate engineering activities (Maas and Scheffran, 2012; Scheffran and Cannaday, 2013; Brzoska et al., 2012; Link et al., 2013). The following discussion distinguishes between five conflict types:
1) Competition over scarce resources: While many albedo modification techniques could likely be implemented with comparatively limited resources, most greenhouse gas removal techniques, such as BECCS or enhanced weathering, demand massive resource inputs to have a globally meaningful impact (see Section 2.1). Large-scale deployment of most techniques for the removal of greenhouse gases would needextensive infrastructures, thereby requiring the extraction and conversion of major resources (energy, raw materials, water, and land) that have an impact on natural and social systems in the affected regions. Thus, competition over physical resources; financial resources like investments; and immaterial resources such as human, social, and political capital could increase, affecting the availability of resources for mitigation and adaptation.
2) Resistance against impacts and risks: Anticipation of foreseeable or suspected impacts; detrimental side effects; and externalities such as pollution, modified rainfall patterns from SAI, or changes in ecosystems, vegetation, and crop yields; as well as principled opposition, might provoke resistance on local, national, and international scales. Furthermore, different techniques for the removal of greenhouse gases have specific local impacts (for example on water, biodiversity, forests, agriculture, or cities) that might encounter resistance from those who are affected and have inadequate coping mechanisms. Moreover, since technical, economic, and political limitations mean that some techniques are feasible only in certain regions, there may be particularly enhanced pressure on the resources and communities in these specific regions. This poses challenges for public acceptance and local coping mechanisms comparable to those associated with other forms of environmental modification, such as large-scale forest clearance for agricultural use, damming rivers, and the creation of artificial lakes (Conca, 2010, Balint, 2011).
3) Conflicts over distribution of benefits, costs, and that the effects of this would differ regionally, there may be international disagreement over what form and scale of SAI deployment (if any) might be considered desirable, as well as over real or perceived injustices in the distribution of impacts from potential deployment. Distributional conflicts may also arise for greenhouse gas removal techniques, especially concerning cost-sharing as well as the distribution of risks of environmental degradation and detrimental impacts on human health or ecosystems.
4) Complex multi-level security dilemmas and conflict constellations: In the absence of international cooperation and broad consensus on high-leverage albedo modification techniques, individual attempts to regulate global mean temperatures could provoke countermeasures by states and the resistance of citizens, leading to potential security dilemmas ranging from local to global levels. In a hypothetical future world in which albedo modification techniques are utilised, it is conceivable that those deploying such techniques might be blamed for weather-related disasters and damage elsewhere, whether justified or not.
5) Power games over climate control: Some have argued that countries may use high-leverage techniques such as SAI as an instrument for power projection and hostile use (Dröge, 2012, Maas and Scheffran, 2012). During the Cold War, the superpowers supported a small amount of research on weather control for offensive and defensive purposes (Fleming, 2010). However, direct military applications of SAI or most other albedo modification techniques appear unlikely for the time being, due to the difficulty of accurately controlling the effects (and even detecting and attributing them). Should it become technologically feasible for countries to attempt to “optimise” their own climate, transboundary effects might trigger diplomatic crises and international disputes that hinder international cooperation.
3.2.2 Economic analysis
Economic analysis of climate engineering is in its infancy and has to be considered in the broader context of climate economics. Most of the focus in climate economics has been on how to control greenhouse gas emissions efficiently, and to explain under what circumstances the costs and benefits of emission control will support “early action” on mitigation. The attention paid by economists to proposals for planetary albedo modification can be traced to a provocativearticle by Thomas Schelling (Schelling, 1996), who pointed out the difficulties in dealing with “something global, intentional, and unnatural” that at the same time has the potential to immensely simplify negotiations over how to address climate change. Schelling argued that albedo modification had the potential to reform climate policy, from an exceedingly complex regulatory regime into a problem of international cost-sharing — a problem with which the world is familiar. Barrett (2008a) subsequently argued that the economics of albedo modification through SAI are “incredible”, representing an opportunity to offset the warming effect of rising greenhouse gas concentrations at a very low cost. However, Barrett (2008a) also highlighted the challenges of governance and regulation (see also Chapter 4). These points of departure may explain why the economic literature on climate engineering has focused mainly on modifying the albedo, especially through SAI, rather than on removing greenhouse gases, which is instead generally discussed within the context of the economics of mitigation. To date, however, economic analyses of albedo modification have been primarily concerned with the possibility of cooling the planet at very low operational cost, often neglecting other costs that this would entail, such as price effects and social costs (see Box 3.6).
The economic literature on climate engineering can be divided into two branches that are further discussed in the subsections below: assessing costs and benefits; socio-economic insights from climate
126.96.36.199 Assessing costs and benefits
Different cost types need to be taken into account when considering the deployment of the various techniques (see Box 3.6 for definitions).
1. Operational costs: cost of installing and maintaining a particular technique at current prices for capital goods and material inputs;
2. Price effects: the effect on prices due to increased demand for certain materials and goods by large-scale implementation of a technique;
3. Social costs: the overall economic cost of deploying a specific technique (i.e., operational costs plus external costs, accounting for price effects).
1) Operational costs: The broad range of estimates for the operational costs of several climate engineering techniques were previously discussed in Chapter 2 and will not be discussed further here.
2) Price effects: Large-scale deployment of greenhouse removal techniques and most proposed albedo modification techniques would generally require large investments, complicated infrastructures, and major material inputs to be effective (Klepper and Rickels, 2012). This may have a strong impact on those markets that provide the sources of such goods and materials. Existing cost studies usually neglect these effects. For example, measures such as spreading lime or iron in the ocean would require a large number of ships, which in turn would lead to a substantial demand shift in the global shipbuilding market (Klepper and Rickels, 2012). Similar effects, although smaller in scale, could be expected for the global airplane market if measures such as SAI were realised on a large scale using aircraft for the injection procedure. Price effects could also occur on the supply side if any techniques were deployed at very large scales: afforestation may increase the supply of wood once the trees mature, and CCS or BECCS may lead to the creation of additional CO2 certificates, affecting the market price of carbon. These various effects would influence relative prices across the entire economy and should therefore be considered when assessing large-scale deployment of individual techniques.
3) Social costs: Despite the uncertainties concerning operational costs and the role of price effects, the greatest uncertainty about the costs of climate engineering, and in particular albedo modification through SAI, arises from its social costs, since present knowledge of such issues is basically non-existent (Klepper and Rickels, 2012). Scientific studies of the various techniques have shown that their use may have unintentional side effects, which can take the form of external costs or external benefits (where “external” in this context implies that a third party not involved in the market transaction has to bear costs or receives benefits). These costs and benefits could be related to the material in use, or to the deployment mechanism. They could also materialise as costs associated with impacts on specific ecosystems or with overall changes in the climate system. For a comprehensive analysis, potential side effects also need to be taken into account and incorporated in the social costs associated with each technique.
For BECCS, external costs are thought to be mainly related to competition for land use and water supply. External costs for OIF would predominantly involve impacts on marine ecosystems. Due to the dynamics of the ocean system, such effects could be distributed widely. Were SAI to successfully reduce global mean temperature, it would change the climate and climate impacts, for example impacting agricultural productivity and the occurrence of extreme weather events. There is limited research on the effect of SAI on various climate impacts, and there is uncertainty over the regional climate response to SAI, which would in turn shape those climate impacts (Robock et al., 2008; Jones et al., 2009; Rasch et al., 2009; Ricke et al., 2010b; Berdahl et al., 2014). However, not all climate impacts are necessarily negative (Klepper and Rickels, 2014). For example, plant water stress is more strongly influenced by the number of extreme hot days than by variations in precipitation (Lobell et al., 2013). For example, high atmospheric CO2 concentration is associated with greater water-use efficiency in some plant species (Keenan et al., 2013); in such a scenario, a reduction in extremely hot days via the deployment of SAI might provide agricultural benefits despite an overall reduction in precipitation, at least compared to an unmitigated climate change scenario (Pongratz et al., 2012). Furthermore, an increase in diffuse irradiation would be expected as a consequence of SAI implementation, which might further promote plant growth (Mercado et al., 2009). However, these considerations of the economic impacts of albedo modification techniques remain very preliminary, since the various interactions and feedbacks are not yet well understood (Klepper and Rickels, 2014) and potential gains in crop yields might only be observed for certain crops in certain regions (Xia et al., 2014).
188.8.131.52 Socio-economic insights from climate engineering scenarios
If one acknowledges the future technological potential for efficient abatement technologies and the slow
transformation of industrial structures, then alternative approaches such as BECCS for removing greenhouse gases might “buy time” for such abatement technologies and transformations to develop (e.g. Kriegler et al., 2013). Nevertheless, it should be borne in mind that the atmosphere is only one reservoir of the global carbon cycle budget. Greenhouse gas removal will cause changes in natural carbon fluxes between the carbon reservoirs, which may significantly impact the effectiveness of the measures, either negatively or positively (e.g. Mueller et al., 200; Vichi et al., 2013; Klepper and Rickels, 2014). Accordingly, in order to properly conduct economic assessments of techniques for greenhouse gas removal and to appropriately model deployment scenarios, studies need to consider various carbon costs, which reflect the social costs that arise from scarcity of storage sites or from the changed ambient carbon fluxes between the atmosphere and other carbon reservoirs (Lafforgue et al., 2008; Rickels and Lontzek, 2012).
Initial findings from modelling exercises and scenario analyses do not yet allow for an overall and comprehensive economic evaluation of SAI deployment or for identifying its potential role in a future portfolio of responses to climate change; however, they do provide a starting point for assessment and important guidance for further research. To date, several studies have already mapped the economic potentials of albedo modification techniques as a policy option, especially SAI (Nordhaus and Boyer, 2000; Bickel and Lane, 2009; Gramstad and Tjötta, 2010; Goes et al., 2011; Moreno-Cruz and Keith, 2013; Aaheim et al., 2015 forthcoming; Bickel, 2013; Bickel and Agrawal, 2013,; Emmerling and Tavoni, 2013); With the exception of Emmerling and Tavoni (2013) and Aaheim et al. (2015 forthcoming), these numerical evaluations employed various versions of William Nordhaus’ Dynamic Integrated Climate-Economy model (DICE; see Nordhaus, 2008). DICE is an integrated assessment model for assessing the costs and benefits of various climate policies. Climate is integrated into the model by linking emissions, via concentrations, to temperature and from there to an aggregated damage function. Climate policies are evaluated by assessing their implementation costs over a given period (usually hundreds of years), compared with the benefits associated with the avoided impacts of climate change. Future costs and benefits are discounted by a chosen rate, which is intended to reflect the return on alternative opportunities for investing the money spent on mitigation. The outcome is critically dependent on assumptions made about the discount rate, which is discussed in further detail in the next section.
The earliest studies aimed to assess the potential of albedo modification by assigning a value for future damages that would be avoided, assuming that the techniques could be implemented at negligible operational costs. Later studies put more emphasis on the influence of side effects (for example, Gramstad and Tjötta, 2010) and in particular on uncertainties associated with implementations of SAI (Goes et al., 2011; Moreno-Cruz and Keith, 2013; Emmerling and Tavoni, 2013). Aaheim et al. (2015 forthcoming) extended these earlier studies by incorporating precipitation changes in assessing the side effects of albedo modification. They employed the GRACE (Global Responses to Anthropogenic Changes in the Environment) model to compare the impacts of cloud whitening and sulphur injection to stabilise global mean temperature over the period 2020 – 2070 in the RCP4.5 pathway scenario. In Aaheim et al. (2015 forthcoming), SAI results in an economic loss globally. This is explained partly by a drier climate, which is expected to result from a combination of SAI with increasing greenhouse gases (see Section 2.2.6), and partly because the SAI simulation misses out on economic benefits that otherwise would have resulted from a moderate increase in temperature in the climate change scenarios without SAI implementation. Nevertheless, the model shows regional variations, with some regions benefiting from the simulated sulphur injection. On the other hand, the results suggested that marine sky brightening would provide an economic benefit in all regions (between 0.1 and 0.8 per cent of GDP). Overall, these studies predict small economic impacts of albedo modification on GDP, but demonstrate that the predicted outcomes are not necessarily beneficial even when the models neglect the operational costs associated with the proposed techniques. However, the model relies on a very simple description of the damages associated with the changes in temperature and precipitation. As discussed in Section 2.2.6, the regional distribution of precipitation changes is not yet well understood, precluding reliable assessment of their economic consequences. Nevertheless, despite these uncertainties, some have argued that the possibility to exert a quick influence on the climate through changes in the albedo allows for greater flexibility in dealing with the uncertainties associated with climate change. Consequently, Moreno-Cruz and Keith (2013) argue that SAI could be a valuable tool to manage risks even if it is relatively ineffective at compensating for CO2-driven climate change, or if its costs are large compared to traditional abatement strategies. Based on this line of argument, they suggest that in any comprehensive risk management approach to climate change, emission reductions and the application of albedo modification techniques should be considered as complementary rather than as substitutes for each other.
3.2.3 Distribution of benefits and costs
The distribution of benefits and costs is not only an economic issue but also raises important normative questions (Burns, 2011). The distributional effects of benefits, burdens, and risks vary considerably between different climate engineering techniques, and therefore need to be discussed individually.
Several authors have argued that SAI would create socalled winners and losers (Caldeira, 2009; Scott, 2012; Barrett et al., 2014), while others have questioned the degree to which SAI would produce inequalities (Moreno-Cruz et al., 2012; Kravitz et al., 2014a). Whether the deployment of SAI would increase the existing inequalities and historical injustice of climate change is an open question. The distribution of benefits and costs would depend not only on existing and uncertain future climate conditions, but also on population density, economic development, and the vulnerability and resilience of ecological, economic, and social systems (Schäfer et al., 2013b). In some scenarios, those geographically and economically most vulnerable to climate change, often living at the subsistence level, would be most likely to be negatively affected by uneven effects of SAI while having the lowest capacity to adapt to such effects, despite being least responsible for global warming (Olson, 2011, SRMGI, 2012; Carr et al., 2013; Preston, 2012). However, others have argued that SAI may also benefit some of the most vulnerable and poorest countries by reducing risks from climate change (Svoboda et al., 2011; Pongratz et al., 2012; Svoboda and Irvine, 2014). The weighing of risks is also an important topic in the context of “lesser evil” arguments, often taken to justify the further engagement in research and the possible deployment of SAI (see Box 3.7).
SAI as the “lesser evil”?
“Lesser evil arguments” are based on the assumption that if no substantial progress on emission reductions is made soon, then at some point in the future a choice would need to be made between allowing certain catastrophic impacts of climate change to occur versus engaging in SAI or another form of albedo modification that might prevent or reduce those impacts but that might also introduce novel risks (Gardiner, 2010; Markusson et al., 2014). In the case of a climate emergency, the argument suggests that the possible negative side effects of a direct intervention in the climate system via a technique such as SAI may be less worrisome than unmitigated climate change (Virgoe, 2009, Preston, 2013; Irvine et al., 2014b). However, in the face of uncertain consequences and unknown side effects of SAI, such claims are debatable. It cannot be ruled out that direct interventions in the climate could worsen some of the harmful consequences of climate change, even if it succeeds in alleviating others (Hegerl and Solomon, 2009, Matthews and Turner, 2009; Rickels et al., 2010).
From an intergenerational justice viewpoint, the distribution of effects of SAI also seems problematic. Based on the assumption that the present generation has a duty to protect the basic interests and rights of future generations (Meyer, 2008), it is widely discussed in the literature that SAI could exacerbate inequalities between generations, as it may allow risks and costs to be deferred (Gardiner, 2010; Burns, 2011; Gardiner, 2011; Goes et al., 2011; Svoboda et al., 2011; Ott, 2012; Smith, 2012). Such deferral of risks and costs would not be unique to climate engineering, as it also applies to the use of fossil fuels and many other activities of modern society. In general, there is a lack of reciprocity between generations of people who are not contemporaries, and an asymmetry in “power”, because current behaviour influences future people whereas they have no possibility to influence the present generation. This can lead to the externalisation of costs and risks over space and time (Ralston, 2009, Gardiner, 2011, Lin, 2012).
In an early analysis, Jamieson (1996) argued that, by deploying SAI, one generation would be choosing a specific climate path for future generations that may be irreversible or only changeable at high costs. Shifting the focus to the generation that would be laying the groundwork for a future SAI implementation
through research and development, Gardiner (2010) argued that it is morally questionable to provide future generations with the possibility of implementing a technique like SAI for their “self-defence” against climate change emergencies and, by doing so, compensating them for a crisis that could have been prevented through more benign options that are still available, such as increased global mitigation efforts (Gardiner, 2010).
In the case of a decision to implement SAI, any subsequent failure to maintain the aerosol forcing to counteract greenhouse gas forcing could result in a rapid and therefore potentially very damaging warming, depending on the scale of the intervention up to that point in time and how abruptly the SAI forcing would be ended (see Section 2.2.7). It has therefore been argued that deploying albedo modification techniques may reduce or foreclose options for future generations to a greater degree than other climate policies (Smith, 2012), thereby impairing or violating their right to autonomous self-determination (Ott, 2012, Smith, 2012) or leading to morally tragic, dilemmatic, or hazardous situations in which agents are compelled to act in a way that is morally reprehensible in at least some sense (Gardiner, 2010, Gardiner, 2011). Generally, intergenerational justice is one of several justice perspectives that can be brought to bear on assessments of the justice aspects of SAI. Others, as discussed by Tuana (2013), include corrective justice, ecological justice, distributive justice, and procedural justice.
3. Emerging societal issues
Similarly to SAI, the environmental side effects of OIF will likely affect large regions, as well as future generations, particularly in terms of irreversible effects on ecosystems and biodiversity. OIF in particular raises questions of ecological justice, in terms of considering adverse effects on non-human life and on ecosystem sustainability in the context of normative evaluations, as well as reflecting upon our responsibilities towards non-human nature (Morrow et al., 2009, Ralston, 2009).
BECCS, despite at first glance perhaps appearing less problematic in terms of distributional effects on the global and intergenerational levels, could still cause harms via land use changes and effects on biodiversity, as a result of the extensive cultivation of monocultures. Due to the need for an adequate feedstock supply, higher levels of deployment would require vast conversion of land. This could decrease the land area available for agriculture and lead to increased food prices.
The potential for some to suffer from the deployment of climate engineering techniques raises questions
concerning compensation for possible harms. Three basic questions can be distinguished for these compensation issues:
Who should compensate?
Whom should they compensate?
What should be compensated?
The question of “who should compensate” links back to the more general debate concerning the main principles for compensation for climate change impacts (Moellendorf, 2002, Page, 2006). The most prominent of these are the “polluter pays” principle (PPP), the “ability to pay” principle (ATP) and the “beneficiary pays” principle (BPP) (Page, 2012). As pointed out by Svoboda and Irvine (2014) as well as Wong et al. (2014), applying one of these principles as the sole governing principle for compensating SAI-induced harms can produce counter-intuitive results. This is often due to the neglect of considerations invoked by the other principles (Wong et al., 2014). Some concerns of applying those principles may be abated if a combination of principles is adopted, as has been suggested more generally for negative effects of anthropogenic climate change (e.g., Page, 2008, Caney, 2010). Combining these guiding principles, especially in relation to potential harms associated with SAI, would not necessarily lead to conflicts between principles, since the principles are not mutually exclusive and can often suggest similar courses of action and similar responsible parties. For example, the countries that would be able to deploy certain kinds of climate engineering techniques over longer periods of time, which would indicate causal responsibility for them in the context of the PPP, would also be those who would most likely gain the greatest benefits, since the form and scale of implementation would tend to reflect their interests (BPP). Furthermore, these would be the countries that would be most able to pay (at least to some extent) for the resulting negative consequences (ATP), and would also be those responsible for most of the historical and/or contemporary emissions (historical responsibility).
It is an open question whether such compensation should be based on the wrongfulness or culpability of
the act, which would place it within the domain of compensatory justice; or based on the need to redress undeserved benefits or harms, which would then be a question of (re)distributive justice (see Box 3.8).
The answer to the question of “who should be compensated” is often less clear than might initially be expected. Different climate engineering techniques may harm different countries in different ways, making them possibly worse off than they would be under global warming alone. The question then is whether all countries would deserve equal compensation, based on the harms caused. Even if different countries faced similar overall losses, there may still be justification for unequal compensation, based on the type of loss, the vulnerability and ability to adapt, the responsibility for global warming, and the gains from it (Bunzl, 2011). Furthermore, an individual nation may simultaneously experience various forms of harm (e.g., increased drought) and benefit (e.g., reduced warming); the balance of these can be very different from one nation to another, adding further complexity to the assessment of who should be compensated.
A third crucial question is “what should be compensated, and to what extent”. Different normative approaches put limits on the kinds of harm that can be compensated. It might be considered impossible to compensate for actions or outcomes that compromise basic human rights or result in the loss of culturally significant ways of life or of statehood. Even for harms that in principle allow for compensation, attributing monetary values may be difficult. The baseline for compensation is also open for debate (Maas and Scheffran, 2012, Preston, 2013, Svoboda and Irvine, 2014). Should compensation claims include adverse effects of past climate change that might worsen or be reduced through climate engineering deployment, or should all compensation claims start from the beginning of the climate engineering activity? Furthermore, on a case-by-case level, it would be challenging to robustly attribute specific harmful impacts, e.g., prolonged drought or flooding, to any form of climate engineering deployment (Allen, 2003; Stott et al., 2004).
Compensation issues are of great importance for SAI. Due to the large effects that an implementation of SAI would be expected to produce, it is likely that not all states would be willing to accept such a course of action without some form of compensation. However, as concluded by Reichwein et al. (2015 in review): “although it is not entirely hopeless, there would be several hurdles in ensuring legal accountability for the risk of environmental harm from SAI under international law”. This is partly because of the inherent nonlinearity and complexity of the climate system, which makes the detection of changes that would be caused by SAI and their attribution to the specific intervention (as opposed to natural variability) highly challenging. This could become less prevalent over time as more data would become available during the decades after deployment (MacMynowski et al., 2011; Jarvis and Leedal, 2012). Additionally, it might be unnecessary to causally attribute an event to some single cause. An alternative option would be to calculate the increased likelihood of an event occurring due to the change in radiative forcing produced by the SAI deployment. However, even calculating the fraction of risk attributable to an event would require comparison of the observed climate with a hypothetical climate in which the climate-forcing activity of interest is excluded; such methodologies would thus rely entirely on model computations, with their associated uncertainties (Stott et al., 2004; Svoboda and Irvine, 2014; Horton et al., 2015; Reichwein et al., 2015 in review).
Compensation issues would also be complex in the case of OIF, and would be concentrated on damages in marine ecosystems, especially in coastal regions. There are various possible impacts for which compensation could be expected, including impacts on the oceanic food web and thus on fish populations and the viability of fisheries (Chisholm et al., 2001), as well as side effects on the atmosphere (Lawrence, 2002) due to various compounds produced by phytoplankton, for instance dimethylsulphide, which can influence aerosol particle concentrations and cloud properties. A further complexity would involve determining who could claim a right to be compensated for damages in international waters. Possible compensation for negative impacts of BECCS would depend on where in the process the negative impacts occur. Questions of compensation could become especially important in the event of possible leakage problems. On the other hand, problems arising during the production of bio-energy could be addressed within existing compensation schemes, as they would most likely occur within the jurisdiction of single nation-states.
4. International regulation and governance
International law frequently uses broad terms to establish the applicability of its provisions. In the issue
area of climate engineering, the term “geoengineering” has become established at the CBD, and “marine geoengineering activities” has become established at the LC/LP. In light of this broad terminology, this chapter does not attempt to differentiate clearly between the regulation of techniques for greenhouse gas removal and for modifying planetary albedo. Instead, the analysis in this chapter suggests the existence of three potential — and partly intertwined — regulatory approaches toward climate engineering, described in Box 4.1, which are slowly becoming apparent in different types of normative output at the international level.
The categorisation of regulatory approaches proposed here serves to illustrate how international and European law could react to the challenges arising out of research on climate engineering techniques and/or their deployment. As such, this categorisation is not explicitly laid down in any binding or non-binding international instrument. That the three approaches discussed below cannot and should not be understood as being clearly distinct from each other becomes evident when taking into account the approach followed by the LC/LP, which is categorised here as being activities-oriented. The LC/LP pursues the aim of protecting the marine environment; any regulatory action taken under its auspices is thus also effect-based. That said, it cannot be denied that the LP, which is set to eventually replace the LC, is based on an entirely different regulatory approach (general prohibition of dumping, with few exceptions) than, for example, the broadly framed CBD. It is asserted that these differences not only legitimise the systematisation introduced here, but also that this systematisation is of key importance for understanding how an effective governance regime covering one or more climate engineering techniques could be designed in future.
Three regulatory approaches for climate engineering
1. regulation of climate engineering based on its potential role as a situational response to various conditions in the overarching context of climate change (context);
2. regulation through risk management measures for individual climate engineering activities and technical processes at the operational level (activities);
3. regulation through scientific assessment of potential environmental effects of different climate engineering techniques (effects). Box 4.1
The analysis then considers how the regulatory approaches surrounding the context, activities, and effects of climate engineering might manifest at the regional level in light of existing sources of EU law, and discusses how the EU has gone about implementing international obligations that fall within these normative categories, which could potentially be applicable to climate engineering. It has already been observed elsewhere that there are shortcomings at the international level in the existing regulatory framework concerning climate engineering (Bodle, 2010; Zedalis, 2010; Rickels et al., 2011; Bodle, 2012; Proelss, 2012; Bodle, 2013). EU law provides, in part, stricter legal standards for environmental protection, and introduces legal innovations that strengthen the regional implementation of global international law and provide a basis for limiting potential climate engineering activities, including unilateral action. An examination of EU law may therefore provide a timely case study of how regulatory structures at the regional/supra-national level might be applicable to climate engineering techniques within a broader framework of multilevel governance.
The remainder of this section (4.1 and its subsections) analyses emerging elements of a potential climate engineering regime in the activities of international treaty bodies. It begins with an overview of relevant treaties and then focuses on three treaties that embody the regulatory approaches outlined above: the UNFCCC, the LC/LP and the CBD. Section 4.2 then contextualises the three regulatory approaches in light of EU law.
4.1 Emerging elements of a potential climate engineering regime in the activities of international treaty bodies
To date, discussion of the development of regulation for climate engineering has primarily taken place in
the competent treaty bodies of the LC/LP and the CBD, although a number of other international treaties, in particular the UNFCCC, would be potentially applicable treaty bodies. Concerning techniques to reflect sunlight back into space, potentially applicable treaties include the Convention on Long-Range
Transboundary Air Pollution (CLRTAP), the Convention on the Prohibition of Military or Any Other Hostile Use of Environmental Modification Techniques (ENMOD), the Vienna Convention for the Protection of the Ozone Layer, and the Outer Space Treaty (OST; for an overview of relevant agreements, see Rickels et al., 2011).
Concerning the ENMOD Convention (which is often mentioned in regard to climate engineering), it was
clearly stated in the “Understandings”, prepared during negotiation of the treaty, that the parties did not
intend its content to be applicable beyond the context of armed conflict. Efforts by the UN Secretary-General to convene a COP at the end of 2013 were also unsuccessful due to a lack of interest among the parties. For these reasons, the ENMOD Convention is not examined further here.
In general, as described in Box 4.1, three nascent and partly interrelated approaches to the regulation of climate engineering are becoming apparent in the activities of the parties associated with the aforementioned treaties. Recapping, these three approaches — along with the treaties that are most closely associated with these approaches in the context of current discussions around regulation of climate engineering — are:
1. context: as a situational, or context-driven, response to climate change (UNFCCC);
2. activities: as an activity or technical process (LC/LP);
3. effects: based on its effects on the environment (CBD).
Note that the first point does not imply that the UNFCCC would have to be considered as an instrument
that has taken account of climate engineering from the outset. The categorisation developed in this chapter solely aims to distinguish between the behavioural patterns on which the different international treaty regimes are based. It has thus been introduced for the sake of systematically approaching the relevant international instruments.
Of course, the aforementioned approaches would not, taken individually, ever be able to provide a comprehensive regulatory framework for climate engineering that would go beyond the regulation of specific climate engineering techniques. Instead, in order to develop an effective regulatory structure for the regulation of climate engineering as such (assuming that such a development would be considered the ultimate objective), all three approaches would arguably have to be integrated. The underlying process could, in principle, occur: formally, at the international level, through a dedicated treaty or protocol; dynamically, by way of, for example, amending the relevant binding instruments in parallel; or informally, through the adoption of assessment frameworks concerning climate engineering research or conclusion of Memoranda of Understanding between the secretariats of the aforementioned treaties. Such joint assessment frameworks and Memoranda of Understanding provide weak forms of formal coupling between the respective treaties at the operational level by fostering the coordinated implementation of shared objectives. Catalytic and synergetic results can be seen in the Rotterdam, Basel, and Stockholm conventions, where treaty integration has virtually been achieved using this approach.
The fact that all relevant legal instruments are to some extent based on similar approaches (embodied,
for example, in the precautionary principle) and thus reflect a common denominator facilitates their integration, whatever form that process might ultimately take. The following subsections discuss the three legal instruments and the regulatory approaches they pursue in their relevance to climate engineering.
4.1.1 UNFCCC — Climate engineering as a context-specific response to climate change?
Identifying climate change as a “common concern of mankind”, the UNFCCC — the most comprehensive regulatory instrument adopted by the international community in response to the challenge of climate change — sets out a very general context into which all efforts to protect the Earth’s climate can be placed. In order to achieve the objective of the Convention, which is set out in Article 2 as the “…stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system”, the UNFCCC provides for two mechanisms:
1. emissions reductions at source;
2. sink enhancement.
These are defined in Articles 1 (8) and 1 (9), respectively. Although no explicit reference is made to climate engineering in the UNFCCC, the definition of 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” arguably covers some greenhouse gas removal techniques and might suggest a potential role for such techniques within the UNFCCC (Proelss, 2012). Other treaty provisions such as Article 4 (1)(c) may also suggest a role for active removal of greenhouse gases from the atmosphere by setting out a duty to “promote and cooperate in the development, application and diffusion, including transfer, of technologies, practices, and processes that control, reduce or prevent anthropogenic emissions of greenhouse gases…”.
Given that the language of the UNFCCC does not provide an explicit legal basis for distinguishing greenhouse gas removal techniques from conventional mitigation, and taking into account that it does not prohibit such activities — but, quite to the contrary, contains references to mechanisms that the academic community now widely understands as a part of greenhouse gas removal — the UNFCCC cannot be interpreted as explicitly prohibiting these forms of climate engineering. At the same time, however, this cannot be expansively interpreted as a blanket authorisation for all greenhouse gas removal techniques.
In any case, the UNFCCC’s formulation of the precautionary principle in Article 3 (3) (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…”) would require a comprehensive assessment of the risks associated with measures to combat climate change (on the basis of the customary duty to inform and the duty to prevent transboundary harm) before a decision on climate engineering research or deployment could be made. This would continue to place primacy on conventional mitigation strategies that emphasise the reduction and prevention of emissions. Further analysis of the relevance of the precautionary principle in the regulation of climate engineering can be found in Proelss (2010), Tedsen and Homann (2013), and Reynolds and Fleurke (2013).
Incentives for the sustainable development of BECCS as a greenhouse gas removal technique under the UNFCCC, should such development be regarded as desirable, would require as a first step that the net impacts of the technology on emissions be recognised in international reporting and accounting frameworks for greenhouse gases. International climate reporting guidelines, as they currently apply to Annex I (industrialised) Parties, only make passing reference to carbon capture and storage (CCS), and accounting guidelines relating to Kyoto Protocol commitments make no mention of CCS at all (IEA, 2011). In 2006, however, a revision of the IPCC guidelines on national greenhouse gas inventory reporting was proposed to recognise CCS. These guidelines make no distinction between CCS using fossil or biomass fuel sources, requiring only that any technology involved in the reduction of emissions satisfies the requirement of the UNFCCC regarding access and technology transfer, that such technologies be “environmentally sound”. The revised guidelines are envisaged to become binding for greenhouse gas reporting from 2015 (IEA, 2011). Further consideration of CCS under the Clean Development Mechanism (CDM) is presented in Box 4.2.
The reporting of biomass related emissions already forms part of the requirements under UNFCCC reporting guidelines. Annex I Parties are required to report such emissions under the LULUCF sector (land use, land use change and forestry). However, while reporting may be comprehensive, the accounting for biomass impacts under the Kyoto Protocol may not be, as parties are able to opt into or out of accounting for certain LULUCF activities. This raises the possibility that the benefits of BECCS in terms of greenhouse gas removal may count toward Kyoto Protocol greenhouse gas commitments, but that the costs of using unsustainable biomass in BECCS may be ignored (IEA, 2011).
Conversely, techniques that aim to reflect sunlight away from Earth are not generally considered to fall within the definitions of “sink” or “emissions reduction at source”, as they do not target the “stabilization of greenhouse gas concentrations in the atmosphere” as required under the objectives of the UNFCCC (Winter, 2011). Potential attempts to subsume such albedo modification measures under adaptation measures rather than mitigation measures, in order to integrate this category of climate engineering into the climate regime, could be challenged on the basis that these activities do not represent conventional adaptation measures such as those provided for in Article 4 (1)(e) of the UNFCCC (“appropriate and integrated plans for coastal zone management, water resources and agriculture, and for the protection and rehabilitation of areas, particularly in Africa, affected by drought and desertification, as well as floods”). Although this list is non-exclusive, there is no indication that a broad interpretation of the term “adaptation” to include albedo modification techniques would be appropriate given the objectives of the UNFCCC contained in Article 2, to stabilise greenhouse gas concentrations in the atmosphere and allow ecosystems to adapt naturally to climate change, because albedo modification techniques fundamentally influence the conditions to which ecosystems would adapt.
4.1.2 LC/LP — Climate engineering as an activity or technical process?
The LC/LP was developed with the primary intention of regulating the dumping of harmful waste and other matter into the oceans. Unlike the UNFCCC and CBD, which enjoy quasi-universal legal status, LC/LP only has a limited international membership, and although most member states of the EU are Parties to the treaty, the EU itself cannot become so as it is not a “State” as required under Article 24 (1). Also in contrast to the UNFCCC, Parties to the LC/LP have actively initiated significant steps towards the regulation of certain ocean-related greenhouse gas removal and sub-seabed storage techniques. Such efforts have concentrated on the development of a risk management framework to regulate potential activities at the operational level, rather than attempting to address the larger contextual questions that climate engineering raises. As such, the LC/LP is a process-oriented instrument which seeks to articulate pathways towards decision-making in situations involving potential pollution of the marine environment, typically through assessment frameworks and amendments to the treaty/protocol. Activity within the LC/ LP COP with relevance for climate engineering was initiated in 2006 in regard to CCS (London Convention and Protocol, 2006), which involved the adoption of an amendment to annex I of the LP to regulate CCS in sub-sea geological formations and an accompanying assessment framework; and a subsequent amendment in 2009 (London Convention and Protocol, 2009), to allow cross-border transport of CO2 for CCS purposes. Note, however, that the introduction of CO2 streams into the water column is prohibited. In the context of climate engineering regulation this development deserves attention, as CCS is regarded as a “bridging technology” when associated with conventionally generated emissions but could arguably be defined as a greenhouse gas removal technique when conducted in direct connection with relevant activities, for example as part of a BECCS approach.
In 2008, the COP adopted a resolution (London Convention and Protocol, 2008) banning OIF (legally defined by Article 184.108.40.206 of the treaty as “placement of other matter”) for purposes other than legitimate scientific research. This was followed in 2010 by an assessment framework (London Convention and Protocol, 2010) through which national authorities can determine whether scientific research on OIF is “legitimate” and how to manage applications to conduct research as required under LP Articles 4.1.2 and 9. A further development, discussed in more detail below, was proposed in 2013. This initial approach to producing non-binding yet politically authoritative guidance for decision making was at the time the most specific regulatory tool regarding any form of climate engineering research or deployment in existence, yet its scope is limited to the marine environment and those atmospheric activities where direct interaction with the marine environment can be reasonably expected. During the course of these and further ongoing developments at the LP, several studies provided insight by examining the question of whether OIF is compatible with international law in general and with the LC/LP in particular (Rayfuse et al., 2008; Craik et al., 2013; Scott, 2013; Verlaan, 2009; Güssow et al., 2010; Markus and Ginzky, 2011).
The most important aspect of the LC/LP process as a potential role model for the future development of governance of climate engineering has been the consideration of risk assessment of potentially polluting activities as a mandatory component of the decisionmaking process, subject to the application of the precautionary principle. The LP mandates a precautionary approach to environmental protection that would be applicable to any form of greenhouse gas removal research or deployment in the marine environment, as well as to the introduction of matter into the marine environment for the purpose of enhancing the planetary albedo, requiring in Article 3 (1) that “appropriate preventative measures are taken when there is reason to believe that wastes or other matter introduced into the marine environment are likely to cause harm even when there is no conclusive evidence to prove a causal relation between inputs and their effects”. Waiving the need to provide conclusive evidence of causation helps to overcome one of the central problems in regulating climate engineering generally — that of scientific uncertainty concerning the effects of research and/or deployment. Under the LP, a state cannot lawfully justify a failure to take preventative measures by reference to a lack of evidence for a causal relationship between an activity and detectable harm to the marine environment. In this regard, the LP’s formulation of the precautionary principle could serve as a helpful interpretive aid within an inter-treaty regulatory structure where formulations of the precautionary principle differ. This is not meant to suggest that the LP’s formulation of the precautionary principle might prevail over other formulations in other treaties. Arguably, this formulation should nonetheless be taken into account as a component of other legal rules applicable between the parties, together with the context, according to Article 31 (3)(c) of the Vienna Convention on the Law of Treaties (VCLT), when interpreting any of the treaties forming part of an inter-treaty regulatory structure. The weighting given to a particular formulation, however, will depend on how the principle is employed in the facilitation of operational cooperation and on the formal recognition of common objectives between the treaties (i.e., within joint COP decisions and Memoranda of Understanding between treaties). Problematic in the case of the LC/LP is its less than universal membership, as noted above. That said, the notion of “global rules and standards” in terms of UNCLOS Article 210 (6)
(United Nations Convention on the Law of the Sea) is often interpreted as referring to the rules and procedures codified in and adopted under the LC/LP, which would result in their incorporation into the regime of that treaty (which is binding upon significantly more states, as well as the EU).
In 2013 an amendment to the LP (LP Resolution 4(8) of 18 October 2013) was proposed by Australia, Nigeria, and the Republic of Korea to extend the scope of the Protocol to regulate the placement of matter for ocean fertilisation and other marine geoengineering activities. This proposal, which may represent a major step forward in the development of regulation for climate engineering, was adopted by consensus in October 2013, and aims to provide a legally binding mechanism to regulate the placement of matter for OIF while at the same time “futureproofing” the LP so as to enable regulation of other marine geoengineering activities that fall within its scope. It constitutes the first binding regulation in international law explicitly referring to a climate engineering activity. However, this statement should not be interpreted as implying that no rules of international law — particularly customary international law and international environmental law — would have been applicable in the absence of this first binding regulation. At the same time, it should be kept in mind that the amendment first enters into force when twothirds of the accepting parties have deposited an instrument of acceptance, and can be subject to reservations, provided these are compatible with the overarching objectives of the treaty. Parties to the LC/LP who do not accept the amendment are not bound by the amendment, but are nonetheless under the obligation to take its provisions into account given the continuing (although legally non-binding) applicability of the resolutions on the regulation of ocean fertilisation (London Convention and Protocol, 2008) and the assessment framework for scientific research involving ocean fertilisation (London Convention and Protocol, 2010) for all parties, in conjunction with the overarching duty to protect and preserve the marine environment derived from the UNCLOS, the LC/LP, and customary international law.
The new Article 1 No. 5 bis contains the first definition of marine geoengineering to be introduced into a
binding international treaty. The amendment furthermore prescribes binding criteria to distinguish research from deployment. The amendment, should it enter into force, would not automatically render the Protocol applicable to all marine greenhouse gas removal techniques. Rather, the applicability of the Protocol depends on a decision by the States Parties to include the activity in question in a new Annex 4 to the Protocol. Concerning techniques that aim to modify the planetary albedo, while the definition of marine geoengineering also encompasses the introduction of matter into the marine environment in order to increase the brightness of clouds, seeding using sea-spray generated from ocean waters would arguably not fall under the regulation. Furthermore, a new Annex 5 transforms the aforementioned Assessment Framework into a legally binding text; it reflects a comparatively strict implementation of the precautionary approach by foreseeing, at several stages of the assessment, that permission should not be granted unless sufficient evidence can be provided that the activity is unlikely to adversely affect the marine environment. Article 6 bis of the Protocol generally prohibits the placement of matter for marine greenhouse gas removal activities unless a listing of the activities concerned provides that they may be authorised under a permit scheme. This process under the LP, of first adopting a non-binding COP decision and then proceeding to amend the treaty/protocol to create binding law, demonstrates a potential model for other legal regimes in regulating other forms of climate engineering research.
4.1.3 CBD — Climate engineering judged in light of its effects on the environment?
The CBD was created with the intended objective to conserve biodiversity. With reference to each state’s sovereign right to sustainably use its natural resources and the duty to prevent transboundary harm as the two foundations for state action, the CBD sets out a regulatory framework to gauge potential harm to biodiversity and ecosystems stemming from human activities. The CBD has a near-universal membership among UN member states, with the sole, but significant, exception of the USA, which has signed but not ratified the treaty. Like the UNFCCC, the text of the CBD itself does not explicitly refer to climate engineering (however, it could be amended by way of a Protocol, cf. CBD Article 28 should the parties decide to regulate climate engineering research and/or deployment in a binding manner). However, in contrast to the UNFCCC, the CBD is not dedicated to the specific context of climate change, and in contrast to the LC/LP it is not designed to regulate specific activities. Its potential role in the regulation of climate engineering is instead to identify normative categories and procedures by which the potential effects of climate engineering on biodiversity can be monitored, assessed, and evaluated, as well as establishing limits, which may not be exceeded, for the reduction or loss of biological diversity.
As set out in Article 3 of the CBD, states have the “responsibility to ensure that activities within their jurisdiction or control do not cause damage to the environment of other States or of areas beyond the limits of national jurisdiction”. To this end, they are required to adopt measures to minimise the adverse impacts on biodiversity, as described in CBD Article 14(1). These measures include the duty to “introduce appropriate procedures requiring environmental impact assessment … and, where appropriate, allow for public participation in such procedures”, as detailed in CBD Article 14 (1)(a). Read in conjunction with the eighth and ninth recitals of the preamble (“Noting that it is vital to anticipate, prevent and attack the causes of significant reduction or loss of biological diversity at source, [n]oting also that where there is a threat of significant reduction or loss of biological diversity, lack of full scientific certainty should not be used as a reason for postponing measures to avoid or minimize such a threat”), the CBD presents an ambiguous version of the precautionary principle that cannot be interpreted as either prohibiting or authorising climate engineering activities.
To date, the CBD COP has adopted two specific decisions explicitly concerning climate engineering (albeit using the term “geoengineering” and without differentiating between albedo modification and greenhouse gas removal techniques), at its tenth meeting in 2010 (Convention on Biological Diversity, 2010) and eleventh meeting in 2012 (Convention on Biological Diversity, 2012). Note that there have also been further CBD COP decisions referring specifically to OIF. The 2010 decision stipulates in Para. 8 (w) that: “…in the absence of science-based, global, transparent and effective control and regulatory mechanisms for geo-engineering, and in accordance with the precautionary approach […] no climate-related geo-engineering activities that may affect biodiversity take place, until there is an adequate scientific basis on which to justify such activities and appropriate consideration of the associated risks for the environment and biodiversity and associated social, economic and cultural impacts, with the exception of small-scale scientific research studies that would be conducted in a controlled setting in accordance with Article 3 of the Convention, and only if they are justified by the need to gather specific scientific data and are subject to a thorough prior assessment of the potential impacts on the environment”.
Although this decision is legally non-binding, it constitutes a politically authoritative statement by the States Parties to the CBD and as such is to be taken into account when measuring climate engineering activities against the biodiversity protection-oriented requirements contained in the CBD.
Decision X/33 also provides a definition of geoengineering that refers to “…any technologies that deliberately reduce solar insolation or increase carbon sequestration from the atmosphere on a large scale that may affect biodiversity (excluding carbon capture and storage from fossil fuels when it captures carbon dioxide before it is released into the atmosphere)…”. It is clear that the CBD considers itself an appropriate treaty body for discussing both greenhouse gas removal and albedo modification techniques, in contrast to the UNFCCC, where the language of the treaty has so far been interpreted as restricting the discussion to greenhouse gas removal techniques (note, however, that the interpretation of the UNFCCC described here might change over time, subject to political will, given the flexible wording of the treaty provisions and the possibility for interpretation of a treaty to change as a result of subsequent agreement between the parties or subsequent practice concerning the application of its provisions, cf. Article 31 (3) lit. a and lit. b of the VCLT, 1155 UNTS 331). Decisions X/33 and XI/20 of the CBD COP are therefore particularly important in relation to albedo modification techniques, as the CBD is arguably at present the only treaty with near-universal legal status in which albedo modification techniques might be considered as falling under its mandate.
Irrespective of the type of climate engineering under consideration, the CBD contains a general obligation to conduct environmental impact assessments (EIAs). The wording of Decision X/33 allows for small-scale scientific research on climate engineering in controlled settings under the condition that these studies are “justified by the need to gather specific scientific data”. Should such experiments take place, the CBD may therefore be under increasing pressure to serve as the framework within which an “adequate scientific basis” can be established “on which to justify such activities”. The nature of justification is different, however, between that required of scientists wishing to conduct small-scale climate engineering research and stakeholders calling for actual deployment.
4.1.4 Outlook: bringing together the regulatory approaches of context, activities and effects
It was noted in the preceding sections that, in order to develop an effective regulatory structure for climate engineering techniques, the three approaches to regulation identified in the activities of the UNFCCC, LC/ LP and CBD would need to be integrated. Whether — and how — this could be achieved depends firstly on better defining the scope of the intended regulation in relation to the specific technical features, areas of application and intentions behind the activities concerned. These specificities have so far not been adequately addressed in international regulatory bodies, mirroring the lack of clarity in the general debate surrounding terms like climate engineering, geoengineering, albedo modification, and greenhouse gas removal (see Section 1.2). Because of the great differences between individual techniques, the prospects and desirability of a treaty that subsumes a wide range of techniques under the general term “climate engineering” and attempts to address the full range of aspects involved (i.e., going beyond specific aspects such as impacts on biodiversity) are clearly negative, taking into account: (1) the time it would take to negotiate such an instrument, (2) that “commons-based” and “territorial” climate engineering techniques raise different jurisdictional issues and would thus require different forms of international cooperation and decision making, and (3) that a clear sense is yet to emerge of what the interests of different actors might be. Shared understandings of technical features, areas of applications and intentions behind climate engineering activities will only emerge — if at all — in light of active, open research programmes and assessments. More fundamentally, the effectiveness of any potential regulatory structure depends on a clear understanding of its object and purpose. It is still premature to speculate on what purpose or purposes the regulation of climate engineering might have and what goals it might be intended to pursue. In the meantime, until a clearer consensus emerges to guide decisions on whether to develop more specific regulation, the alternative is to focus on bringing together the aforementioned regulatory approaches at the operational level (i.e., through parallel action, common assessment frameworks or Memoranda of Understanding) using the example of those climate engineering techniques that are most relevant to research and practice. In this respect, developments that have taken place within the legal framework of the LC/LP might serve as a model for other non-ocean-related climate engineering activities.
4.2 The EU law perspective: considering a potential regulatory strategy for climate engineering including application of the approaches of context, activities and effects
Due to climate engineering’s inherent potential for significant transboundary environmental effects, the question of its overarching permissibility remains at the level of public international law. An exclusively “top-down” perspective, however, would fail to take into account the more heterogeneous processes by which international law is created and implemented, as well as the different yet interlinked normative planes necessary for a coherent and stable legal system. Regulatory structures that would be initially applicable to climate engineering research or deployment activities already exist at the national and supranational levels, and these could provide a structure for implementing a new international legal regime within the dynamics of multilevel governance. Without a comprehensive international regulatory structure in place, EU law provides a “bottom-up” source of limitation on climate engineering for member states and the EU itself. Although present EU law (the scope of which is limited to the territory of EU member states and to activities undertaken by EU citizens abroad) cannot be interpreted as generally prohibiting or authorising climate engineering, it serves to structure the decision-making process and provide essential provisions for environmental protection, which are in any case required to be implemented and enforced. EU law consists of both primary and secondary sources, with the former consisting of the multilateral treaties adopted by the member states defining the objectives and competences of the EU as an institution, and the latter consisting of the supranational legal mechanisms created by the EU itself for realising those objectives. In some cases such as the UNFCCC and CBD, the EU is also a party to multilateral treaties alongside its individual EU member states, which provides a further source of secondary law for the realisation of EU objectives. In other cases such as the LC/ LP, the EU is not a party, meaning that member states are exclusively and independently responsible for implementing their obligations under this treaty, something which must nonetheless take place in compatibility with the framework of EU law. The EU is generally not precluded from enacting compatible (or even stricter) internal regulations, provided that the subject matter falls within its competences. However, it cannot exercise a formal coordinating role among the EU member states’ positions within treaty bodies to which it is not a party. This fact considerably complicates the formation of a common EU position on a given topic “externally”, within the sphere of international relations, and therefore the influence that such a common position might have for the evolution of the treaty. This complex and continually evolving legal landscape on the one hand complicates an assessment of existing and emerging legal instruments at the international level, but on the other hand opens the analysis of new sources of law and provides an additional normative forum within which climate engineering can be subjected to more democratically legitimated legal scrutiny, at least within the EU, than at the level of international law alone.
4.2.1 EU Primary Law — An overarching context for climate engineering regulation and competences for its implementation within the EU
Questions regarding a regulatory strategy to determine a potential context within which research or deployment of climate engineering could be authorised or prohibited in the EU are best posed at the level of EU competences and objectives, i.e., against the provisions of EU primary law. The Treaty on European Union (TEU) (The Member States of the European Union, 2012a) sets out the organisational structure and objectives of the European Union that are relevant for potential EU action in the field of climate engineering. The Treaty on the Functioning of the European Union (TFEU) (The Member States of the European Union, 2012b) sets out in more detail the nature of competences between the Union and the member states, including the environment, energy, and common safety under TFEU Article 4 (2). TFEU Article 4 (3) adds that “[i]n areas of research, technological development and space, the Union shall have competence to carry out activities, in particular to define and implement programmes; however, the exercise of that competence shall not result in Member States being prevented from exercising theirs”. Because action concerning environmental protection on the one hand and research on the other hand is subject to shared competences under EU law, both national and supranational activity in climate engineering research and deployment are possible at present. As evidenced by the developments concerning CCS, this could potentially mean that both the EU and some member states engage in research and technological development of climate engineering methods, while other member states, despite not being in the position to veto EU-level research and development following the adoption of the CCS Directive, find themselves unable to promote such activities domestically due to public opposition of the kind witnessed in Germany over CCS (for an overview on the state of implementation of the CCS Directive and the divergent positions taken by EU member states, see COM(2014) 99 final). With regard to environmental competence, it should be noted, though, that according to TFEU Article 2 (2) the member states may only “exercise their competence to the extent that the Union has not exercised its competence. The Member States shall again exercise their competence to the extent that the Union has decided to cease exercising its competence”. Were the EU to enact a moratorium, prohibition or authorisation of climate engineering in general or for individual climate engineering techniques, this would either need to be set out in an international treaty to which it is a party, which would then be transposed into EU secondary law, or by making use of its environmental competence codified in TFEU Article 192 (2), on the basis of a Commission proposal through the standard legislative process.
Regarding the Union policy on the environment, TFEU Article 191 (1) sets out the primary objectives, including “preserving, protecting and improving the quality of the environment, protecting human health, prudent and rational utilisation of natural resources, promoting measures at the international level to deal with regional or worldwide environmental problems, and in particular combating climate change”. As the central guiding provision of EU primary law with relevance for climate engineering, TFEU Article 191 (2) expressly requires that environmental policy “shall be based on the precautionary principle and on the principles that preventive action should be taken, that environmental damages should as a priority be rectified at source and that the polluter should pay”. On this basis, all climate engineering activities would be judged against the precautionary principle as a binding rule of law, as well as the duty of preventive action against emissions and the mitigation of emissions at source, as opposed to either their dispersal or sequestration via other environmental media as intended by greenhouse gas removal techniques, or the management of solar radiation as targeted by albedo modification techniques. In accordance with TFEU Article 191 (3), the Union is also required to take account of available scientific and technical data in preparing environmental policy, as well as of the potential benefits and costs of action or lack of action, taking into account the enormous degree of scientific uncertainty concerning climate engineering.
Given these principles and standards of care for interaction with the environment, it is unlikely that most climate engineering methods would satisfy the principle of preventive and precautionary action. At the same time, as far as greenhouse gas removal is concerned, the distance between the source of emissions and their point of sequestration would be a decisive factor in determining whether a certain activity can be seen as sufficiently rectifying emissions “at source” as required by EU law. Finally, the overarching objective of EU environmental policy of “improving” the quality of the environment and rationally using natural resources could be threatened by the prospect of climate engineering. Despite the prominent position given to combating climate change within EU environmental objectives and the clear intention behind climate engineering to contribute to such ends, in order to be consistent with EU policy climate engineering techniques would nonetheless need to be judged on their capacity to satisfy all objectives of EU environmental policy simultaneously, rather than merely as a second-order measure to satisfy one individual objective.
4.2.2 EU Secondary Law
Hier weiterlesen: Seite 2