Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CLIMATE SCIENCE ( (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 08 August 2020

Climate Engineering

Summary and Keywords

The means to combat dangerous anthropogenic climate change constitutes a portfolio. Beside abatement of greenhouse gas emissions, this portfolio entails adaptation to changing climate conditions, and so-called climate engineering measures. The overall portfolio has to be judged on technical, economic, and moral grounds. This requires an in-depth understanding of the moral aspects of climate engineering options. Climate engineering (CE) is a large-scale intentional intervention either in carbon cycles (carbon dioxide removal; CDR) or in solar radiation (solar radiation management; SRM). The ethical discourse on climate engineering has gained momentum since the 2010s. The set of arguments pro and contra specific CE technologies constitute a vast landscape of discourse. Single arguments must be analyzed with scrutiny according to their ethical background, their normative premises, their inferential logic, and their practical and political consequences. CE ethics, then, has a threefold task: (a) it must suppose a solid understanding of different CE technologies and their risks; (b) it has to analyze the moral arguments that speak in favor or against specific CE technologies; and (c) it has to assess the impacts of accepting or rejecting specific arguments for the overall climate portfolio’s design. The global climate portfolio differs from ordinary investment portfolios since stakes are huge, moral values in dispute, risks and uncertainties pervasive, and collective decision-making urgent. Any argument has implications of how to design the overall portfolio best. From an ethical perspective, however, one must reflect upon the premises and inferential structures of the arguments as such. Analysis of arguments and mapping them logically can be seen as core business of CE ethics. Highly general arguments about CE usually fall short, since the diverse features of individual technologies may not be addressed by overarching arguments that necessarily homogenize different technologies. It can be stated with confidence that the moral profiles of CDR and SRM are highly different. Every single deployment scheme ought to be judged specifically, for it is a huge difference to propose SRM as a substitute for abatement, or to embed it within a comprehensive climate portfolio including abatement and adaptation, where SRM will be used sporadically and only for a matter of decades.

Keywords: climate ethics, climate engineering, carbon dioxide removal, solar radiation management, natural climate solutions, moral hazard, buying time, political economy

The Debate on Climate Engineering in the Context of Climate Change

Climate Change and the Paris Agreement

In November 2015, the 21st Conference of the Parties (COP21) to the United Nations Framework Convention on Climate Change (UNFCCC) took place in Paris. All 197 participating states mutually agreed on the danger and urgency of climate change being the foremost threat to human society. The final Paris Agreement aims at reducing increase in global mean temperature to “well below 2°C above pre-industrial levels” (UNFCCC, 2016, p. 3). In order to reach this goal, global greenhouse gas (GHG) emissions should reach net zero by the end of the 21st century, asking for drastic emission cuts (abatement) in the coming decades for most countries, except the least developed countries, but including the BRIICS states (Brazil, Russia, India, Indonesia, China, South Africa). The Paris Agreement marks the preliminary end of the long debate on anthropogenic climate change, setting a reasonable objective that is in line with Article 2 of UNFCCC.

The nationally determined contribution (NDC), which each country formulates individually, is a nation’s share to realizing this supreme goal. The NDCs are a crucial mechanism under the Paris Agreement. They are expected to amplify every five years, and the nations are expected to eventually abandon the use of fossil fuels (UNFCCC, 2016, p. 3). By April 2019, 184 out of 197 parties, plus the EU, have ratified the commitment—with the exception of the United States, which announced its intention to withdraw from the Paris Agreement in June 2017.

Because the NDCs will be expected to become more ambitious over time, it is not in a state’s interest to define ambitious NDCs in the first instance. Thus, existing pledges are not sufficient to reach the well-below 2°C target.Many NDCs made by countries of the Global South are conditional on proper adaptation financing. Moreover, the mechanism to implement the NDCs is argued to be feeble, and the power of the treaty is supposed to rely mainly on a country’s good reputation (Jacquet & Jamieson, 2016), since there are no sanctions.

A grand, large-scale, and global effort to reduce CO2 emission is needed in order to meet the commonly agreed climate targets (Baatz & Ott, 2017a). (There are several other greenhouse gases (GHGs) that contribute to global warming. Because the main driver of global warming is CO2, the article will solely focus on CO2.) The first signs of climate change are already visible (IPCC, 2014b), some of which may be irreversible (Solomon, Plattner, Knutti, & Friedlingstein, 2009) and both human and natural systems suffer from the impacts. The threat of a “climate emergency” looms behind climate predictions, while scientists repeatedly correct their predictions to ever more dire scenarios (e.g., Hansen et al., 2016).

It is in this specific situation that the idea of climate engineering (also known as geo-engineering, climate manipulation, or climate intervention, hereafter referred to as CE) arose. In Fleming (2010), the ideas of engineering the climate go back to former projects of weather manipulation. Early ideas of solar radiation management have been presented in Teller, Hyde, and Wood (2002). Some years after Teller et al.’s rough calculations and moral affirmation on solar radiation management, Crutzen (2006, p. 217) opened up the discussion with his editorial remark, urging for “active scientific research” on stratospheric aerosol injection (SAI)—a technology that may possibly reduce radiative forcing by injecting sulfate particles into the stratosphere (overview in Niemeier & Tilmes, 2017; for some technical details, see Niemeier, Schmidt, & Timmreck, 2011). Crutzen’s article was soon followed by Victor (2008) and Victor et al. (2009, p. 76) in the affirmative: “It is time to take geoengineering out of the closet.” Robock (2008) presented a list of 20 arguments why geo-engineering may be a bad idea. Since the 2010s, the debate on CE has become highly intense.

On Terminology

Climate engineering has been defined as the “deliberate large-scale manipulation of the planetary environment to counteract anthropogenic climate change” (Royal Society, 2009, p. 77). Deployment of CE technologies may avert dangerous climate change or its impacts in the face of insufficient mitigation. And with the fifth assessment report of the Intergovernmental Panel on Climate Change (IPCC, 2013) having included CE as an (admittedly theoretical) option to lessen climate change impacts, CE has been hoisted into the realm of acceptable climate policy options.

Climate policy options (or portfolios) contain different strategies on how to cope with climate change and its impacts. They include options to reduce GHG emission (abatement), adjust to climate impacts (adaptation) and, as of late, technically averting climate effects. The concept of mitigation has several meanings. While some use it synonymously with abatement, the IPCC (2014a, p. 4) defines mitigation as “a human intervention to reduce the sources or enhance the sinks of greenhouse gases.” This official IPCC definition, which correlates with the definition of mitigation listed in UNFCCC (2016, Article 4), mentions two aspects of mitigation: reduce the sources and enhance the sinks of GHGs. According to this definition, CDR measures can be seen as part of mitigation. Especially bioenergy and carbon capture and sequestration (BECCS) aim at capturing CO2 before it enters the atmosphere, and hence can be said to reduce the source of emission. Additionally, biomass acts as a natural sink for CO2. Natural climate solutions (NCS) would also count as mitigation. By definition, the line between reduction of emissions (abatement) and CDR is blurred. The following adopts the definition out of IPCC’s authority, but it remains open for debate, whether this definition is helpful.

Alternative terminologies have been proposed in Heyward (2013) and Boucher et al. (2014). Heyward (2013, p. 25) suggests to distinguish strategies to combat climate change according to the point at which they occur in the “process between emitting GHGs and the loss of human wellbeing.”

Table 1 . Categorization of CE Technologies According to Heyward (2013).


Avoiding climate change

Avoiding dangerous climate change

Responding to dangerous climate change

Avoiding a given level of atmospheric GHG concentration.

Avoiding global average temperature increases.

Ensuring that rising temperatures do not impact upon core interests.

Providing redress for injuries to core interests.



Reducing GHG emissions.


Drawing GHGs out of the atmosphere.


Increasing Albedo.


Improved irrigation, flood defences, protection against disease.


Financial compensation, symbolic reparation.

In this categorization, mitigation and CDR are distinctly separated, as the former aims at reducing CO2 output in the first place, and the latter at minimizing any given CO2 concentration ex post.

Boucher et al. (2014) argue that an unclear definition of CE technologies may obscure and even prevent political decision-making. They propose a different terminology, in which CDR measures no longer count as mitigation measures, but rather constitute a separate form of climate interventions, and SRM technologies may even fall under adaptation measures. Most notably, they present a separate category of domestic CO2 removal, which enables the distinction between local and global CDR.

Table 2. Incomplete Table According to Boucher et al. (2014).

Proposed Name and Acronym

Short definition


Anthropogenic emissions reductions (AER)

Initiatives and measures to reduce or prevent anthropogenic emissions of warming agents into the atmosphere.

Improved energy efficiency, reduction in production and/or consumption of goods and services, introduction of renewable energies, nuclear energy, fossil fuel energy with CCS, reducing emissions from deforestation and forest degradation, emission reductions of BC and ozone precursors.

Territorial or domestic removal of atmospheric CO2 and other long-lived greenhouse gases (D-GGR)

Removal of CO2 and long-lived greenhouse gases from the atmosphere operating within national jurisdictions and little consequences outside.

Reforestation, biochar and other means of increasing storage of C in soils, small-scale afforestation, BECCS, CO2 air capture and storage in territorial (geological) reservoirs, enhanced weathering (without input of by-products into rivers or the oceans).

Trans-territorial or trans-boundary removal of atmospheric CO2 and other long-lived greenhouse gases (T-GGR)

Removal of atmospheric CO2 and long-lived greenhouse gases from the atmosphere operating or having consequences partly or fully across or beyond national jurisdictions.

Large-scale afforestation, ocean alkalinity, enhanced weathering (with input of by-products into rivers or the oceans), iron fertilization, injection of CO2 into the ocean.

Regional to planetary targeted climate or environmental modification (TCM)

Intentional modification of the Earth’s energy fluxes in order to offset climate change at the regional to global scale.

Injection of stratospheric aerosols, marine cloud brightening, cirrus suppression, desert brightening on a large scale, ocean heat mixing, modification to Arctic sea ice.

Climate change adaptation measures including local targeted climate or environmental modification (CCAM).

Initiatives and measures to reduce the vulnerability of natural and human systems to the effects of climate change. Local risk management.

Relocating urban or rural settlements, building dykes, air conditioning, agricultural crop choices, reflective crops, whitening of human settlements on a small scale, irrigation.

Only the definition and examples are reprinted, while the original table of Boucher et al. also includes: mapping onto existing terminology, scale of action, scale of impact, impact on the global commons, trans-boundary or trans-national side effect and permanence of the effect.

Terminology should follow the epistemic maxim: Keep it stupidly simple! (KISS). This article relies on the more conventional understanding of curbing emissions (abatement), adaptation, and removing carbon from atmosphere (CDR) and changing solar radiation (SRM) as two large types of CE. The portfolio of strategies includes these four types of assets.

CE aims at delaying or even offsetting climate change impacts technically by either manipulating the global mean temperature directly (SRM) or by removing CO2 from the atmosphere (CDR). The moral profiles of SRM and CDR greatly differ. These differences imply that a general moral or political judgment on CE is impossible. Judgment on CE as such is neither right nor wrong, but pointless. Any such judgments must specifically refer to either SRM or CDR or to a mixture of them, or a combination of CE with abatement and/or adaptation, because SRM and different DCR technologies have different moral profiles and require more fine-grained moral, economic, and political assessment and judgment. Thus, SRM and CDR are two different classes of assets in the portfolio which require further specification. Figure 1 visualizes the CE options.

Climate Engineering

Figure 1. Different climate engineering technologies, sorted by SRM and CDR approaches (Rita Erven, GEOMAR, 2019).

However, the endeavor of CE comes with an array of problems—technologically, politically, and morally. With research on CE still being in its infancy, the scope and effectiveness of CE is but a possibility. There are grave, maybe even insurmountable uncertainties about the side effects of CE deployment, since both the technologies as well as the climate system are not sufficiently understood yet. Social impacts of developing and advertising CE technologies are uncertain. There is no viable political governance framework that could enable a global governance of CE technologies with respect to legitimacy, liability, or compensation. The debate about CE faces incomplete and uncertain information impeding robust decision-making. Since the 2010s, a multitude of proposals have been made with respect to SAI, BECCS, ocean upwelling, enhanced weathering, cloud seeding, reforestation and afforestation, and so-called “natural climate solutions,” which search for win-win solutions between nature conservation and mitigation. The article at hand gives an overview and it conceives a general ethical method of how to reason about CE.

On Method: Portfolio Perspective and Maps of Argument

It seems convenient to combine two methodical approaches: the portfolio perspective, and the maps of CE arguments being designed by Betz and Cacean (2012). There are other approaches to map the ethical arguments on CE that differ from Betz and Cacean (2012). Preston (2013) and Ott (2011) have also summarized and categorized the central arguments. The portfolio perspective presents different assets (means, strategies) to combat climate change and its impacts while the map of arguments presents prudential, political, and moral reasons by which an asset or a combination of assets may be judged. The portfolio perspective is descriptive, the map of arguments is a broad source of normativity. The combination of both enables participants in the CE debate to conceive arguments and judgments. An instance of how to do so is to be found in Neuber (2018) with respect to the buying time argument. The map of arguments is dynamic. Some arguments that played a prominent role in the first exploratory phase of the discourse must be revised and refined (Reynolds, Parker, & Irvine, 2016).

The best way to manage climate change is arguably through a portfolio of different strategies. All CE assets, then, may be part of a comprehensive climate policy portfolio, adding to mitigation and adaptation (Morrow, 2014; Rayner et al., 2013; Wagner & Zeckhauser, 2012). However, the portfolio approach to climate politics may oversimplify the problems at stake. Since CE and mitigation may have interdependencies, and CE may inflict a reduction in mitigation efforts, a purely additive concept of a portfolio may be myopic (Gardiner, 2013; Morrow, 2014). This portfolio perspective may also suggest that economic considerations (cost efficiency) should prevail in determining the efficient response to climate change. Climate ethicists have emphasized that moral reasoning should be intrinsic to the portfolio debate. It may make a difference whether a means (asset in the portfolio) cures the symptoms or addresses the root cause of climate change. The global climate portfolio differs from ordinary investment portfolios since stakes are huge, moral values in dispute, risks and uncertainties pervasive, and collective decision-making urgent. The figure of a benevolent portfolio manager is discarded, and a discourse ethical approach is adopted instead. Any argument has implications of how to design the overall portfolio best. From an ethical perspective, however, one must reflect upon the premises and inferential structures of the arguments as such.

Under this approach, the portfolio approach should be combined (or fused) with the map of pro and con arguments by way of hypothetical reasoning. To contribute to the CE debate is to substantiate validity claims of the following structure: “Asset A should (or should not) be part of the overall portfolio because of the following reasons (R1, R2, …, Rn).” This structure is dialectical, because any validity claim strives for agreement, while the plurality of both assets and reasons seem to make agreement on the best portfolio an instance of wishful thinking. Figure 2 presents a simplified idea of the portfolio.

Climate Engineering

Figure 2. Three possible strategies in a climate policy portfolio: mitigation, adaptation and climate engineering. For the latter some areas of research are indicated (Rita Erven, GEOMAR, 2019).

Ever since climate change was recognized as a global problem, responses to this problem were categorized as being part of either a mitigation or an adaptation strategy. The IPCC 2007 assessment report knows only those climate strategies, as well. IPCC (2007, p. 56) states that “societies can respond to climate change by adapting to its impacts and by reducing GHG emissions (mitigation).” Mitigation efforts can be determined in resemblance to the four representative emission pathways (RCPs) that have been modeled to illustrate emission trajectories and corresponding effects (van Vuuren et al., 2011). RCP2.6, for example, is deemed to be coherent with limiting global temperature rise to (less than) 2°C above preindustrial level. Since both political development between the IPCCs first release and the early 21st century as well as natural inertia (many effects of past emissions cannot be reversed; Solomon et al., 2009) give rise to the belief that limiting global warming by aggressive and far-reaching abatement of emissions only is out of reach. The decades between 1990 and 2020 are a period in history in which both awareness of climate change as a global problem and GHG emissions increased due to a rapidly globalizing economy. Therefore, debate on assets beyond abatement has become unavoidable (McNutt et al., 2015a, p. 18).

Consequently, the need for other, additional solutions is voiced. Since the article focuses on CE, it does not address the highly complex topic of adaptation (see Adger, Lorenzoni, & O’Brien, 2009; Pelling, 2011) and adaptation finance (Baatz, 2018). The most contested set of technologies to combat global warming is known as climate engineering. CE has been openly discussed as a third option besides mitigation and adaptation in Crutzen (2006). It has since been advocated as part of a reasonable climate policy portfolio (e.g., Keith, 2013; Long, 2016).

On Climate Engineering Ethics

Ethics is a theoretical reflection upon moral discourse. The core of discourse ethics is to clarify the commitments involved in the very practice of reasoning (Ott, 2017) and, more substantially, to reconstruct patterns of reasoning being used in specific fields of discourse (as climate ethics and CE). Most substantial arguments must rest on contested premises. This makes CE discourse an exercise in hypothetical reasoning (if-then). A seminal analysis of the iffy logic of hypothetical reasoning has been given by Rescher (1964).

A course of action A can be either obligatory (mandatory), permissible (allowed), or forbidden. Permissions can be either conditional or unconditional. Moreover, a person P may have a moral or legal right (entitlement) to perform A. The right to perform research may entitle scientists to research SAI. States may claim rights to unilateral CE actions. Rights may be claimed and denied. Any moral argument on CE must state whether a course of action should be permissible, obligatory, or forbidden on moral grounds.

On a more general note, many philosophers have argued that the enterprise of technically counteracting climate change may be as such (intrinsically) morally wrong (Gardiner, 2010, 2013; Robock, 2008). They rely on the ethical distinction between actions that are intrinsically wrong and actions that are wrong out of their (likely) consequences. A wrongful action is prima facie (unless there is another overriding moral reason) forbidden. Even if, for instance, SAI may not be wrong intrinsically, it may be wrong if there is the likeliness of a moral hazard. Some authors argue that pursuing drastic emission cuts is the foremost priority of developed nations, and CE may divert from this obligation.

It seems appropriate to distinguish research and deployment of CE as two courses of action that may, of course, overlap, as in the case of large field trials. It is common wisdom that it seems harder to argue against CE research than against CE deployment. Research may be mandatory in order to make the assets within the portfolio real (“arm the future” argument). If, however, deployment of specific CE measures is clearly forbidden on moral grounds, it seems wasteful (inefficient) to invest research energies into such fields of inquiry.

A Look Upon Climate Engineering Technologies

It is widely agreed in literature that CE technologies can broadly be categorized into solar radiation management (SRM) and carbon dioxide removal (CDR), containing different assets. These assets may or may not become part of the overall climate portfolio. CE ethics, then, has a threefold task: (a) it must suppose a solid understanding of different CE technologies and their risks; (b) analyze the moral arguments that speak in favor or against specific CE technologies; and (c) assess the impacts of accepting or rejecting specific arguments for the overall climate portfolio’s design. Analysis of arguments and mapping them logically can be seen as core business of CE ethics.

Solar Radiation Management

In order to control and stabilize the global mean temperature, SRM technologies directly influence the energy balance of the Earth by reflecting the incoming sunlight and thus influence the radiative forcing. Reflecting SRM schemes may be situated in the atmosphere (reflective aerosols) or even in outer space (space mirrors). While the idea of space mirrors would actually amount to reflective particles orbiting Earth, this idea is not very promising due to technical and financial difficulties. Deployment costs would be in the magnitude of some trillions.

The idea of “whitening” areas of land in order to change the albedo has been discarded due to lack of effectiveness (see McNutt et al., 2015a, p. 128f). Some more promising strategies include the enhancement of cloud formation (cloud seeding) over the ocean, which alters the albedo and shields a significant amount of sunlight. However, the physics behind cloud formation is still imperfectly understood, which may lead to great uncertainty regarding the side effects of this set of technologies (Royal Society, 2009). Another idea to modify clouds refers to Arctic cirrus clouds (Lohmann & Gasparini, 2017), but the effects of cirrus cloud seeding seem limited (Gasparini & Lohmann, 2016).

Injection of reflective aerosols, mostly sulfate aerosol (SAI), has been researched and debated at large. There is a broad recent corpus of literature focusing on different aspects of this technology, ranging from the technical and physical aspects over calculations of deployment costs to moral issues and governmental challenges. The latest comprehensive overview on the different aspects of this technology can be found in McNutt et al. (2015a, 2015b). It summarizes the state of the art on SAI research. SAI is the most tempting SRM technology, because it is technologically feasible, deployment cost look cheap, and it brings about quick effects.

SAI technologies, however, do not address the root cause of climate change—the concentration of CO2 in the atmosphere. This is a reason for a mostly negative stance toward SAI technologies, since a symptomatic approach toward climate change is associated with the reprehensive attitude of techno-fixing climate change. Even proponents of SAI agree that there are repugnant initial intuitions and emotions against SAI deployment, but argue that one should not ground a discourse on SAI on intuitions and emotions.

Carbon Dioxide Removal

The second group of climate engineering strategies aims at removing carbon dioxide from the atmosphere. This may happen through mechanical or technical carbon air capture, for example via artificial trees, or by enhancing natural CO2 sinks, such as enhanced weathering, ocean fertilization, restoration of mires to make peat layers grow, enhancement of oceans alkalinity, and reforestation and afforestation. Even the differences between reforestation of land, where forests were converted into grazing or arable land, and large-scale afforestation of deserts are substantial. Reforestation induces conflicts over competing land use. Large-scale afforestation requires large amount of freshwater, and it faces a trade-off between carbon removal and albedo change (Rickels et al., 2011; Royal Society, 2009). With respect to forests, assisted reforestation that generates income to local people should have priority over large-scale afforestation. New studies indicate that restoring natural forests can contribute substantially to CDR (Lewis, Wheeler, Mitchard, & Koch, 2019).

Some CDR technologies have been classified as natural climate solutions (NCS), as they offer win-win situations with goals of biodiversity protection, ecosystem restoration, soil formation, and the maintenance of different kinds of so-called ecosystem services, as providing, regulating, and cultural services. Biodiversity and ecosystem service objectives may be reached by means of preservation, cultivation, and restoration. NCS are interferences that conjoin C-management with such objectives. Important ecosystems are mires, forests, coastal zones, and soils. Mires store large stocks of carbon within peat. Enhancing peat formation in mires counts as CDR and NCS. Hans Joosten’s group in mire ecology has argued that strategies for mires combines cultivation (paludiculture), CDR, and nature conservation are highly promising. If applied at once the release of methane will be over before 2070 and carbon removal brings positive net results (Joosten, oral communication, May 2019). C-enhancement of soils is also a promising strategy. Fertility of soils is crucial for global food security. NCS should be financed by the adaptation financing schemes. Some models, however, indicate that NCS will not bring global temperatures well below 2°C unless reduction of emissions would be dramatically increased. NCS face limitations in scope and in effectiveness (Anderson et al., 2019). They take time to show effect and require the virtue of patience. Nevertheless, they have found growing attention in the literature because their potentials may have been underrated (Griscom et al., 2017). NCS, however, face the risk that carbon being stored may be released in the future since organic nature can only provide storage, not final sinks. It may be argued, however, that the societies that have performed climate policies over decades will not be so foolish as to remove new-grown forests or drain restored peatlands.

CDR modeling has focused on technologies that combine bioenergy with carbon capture and storage (BECCS). BECCS means that biomass is harvested and combusted for energy production, while the resulting CO2 emission is captured and stored underground. BECCS is seen as a very promising approach in climate policies. Most models that reach the Paris target (with a likeliness of more than 66%), strongly rely on negative emissions after 2050 being produced mainly by BECCS. BECCS, however, if performed at a Gt scale, requires large amounts of fertile land for biomass production (Creutzig, 2014). Thus, it conflicts with food security of a global population of roughly 9.6 billion humans in 2050. Locations for BECCS should be the tropics and subtropics where population increase is high. It matters in terms of yields, whether biomass will be produced with or without irrigation. Yields are far higher with irrigation but this may enhance water scarcity and, thus, constitutes a harsh trade-off (Bonsch et al., 2016). Relying on future BECCS in climate policymaking in the early 21st century transfers grave risks into the future. According to Anderson and Peters (2016, p. 183), outlook for large-scale BECCS are an “unjust and high-stake gamble.”

All CDR strategies are contested with respect to costs, conflicts, and removal capacities. Even though CDR technologies treat the root cause of climate change, such as the concentration of carbon dioxide in the atmosphere, they too have by and large great influence on natural cycles. Some CDR measures, such as direct air capture, may not negatively affect the carbon cycle, though. Other CDR technologies may fall under international regimes on nature conservation, as Convention on Biological Diversity (CBD), or Ramsar Convention (Armeni & Redgwell, 2015; Reynolds, 2014a, 2015). Besides these problems, CDR technologies are generally more expensive than SRM schemes, and they take longer to show effect. As such, they may not be able to prevent a sudden and catastrophic change in climate.

Overview on the Moral Debate About Climate Engineering

The first comprehensive argument analysis on the moral controversy about CE has been conducted by Betz and Cacean (2012; see also Preston, 2013; Rickels et al., 2011). In their landmark report, Betz and Cacean map the normative issues surrounding CE. They distinguish arguments regarding the future deployment and those regarding the research of CE technologies. Moreover, they use the placeholder T for any specific CE technology. This would resemble the placeholder A for asset in the portfolio perspective. Betz and Cacean argue that some arguments only apply to a certain technology or a group of technologies. Arguments about CE in general usually fall short, since the diverse features of individual technologies may not be addressed by overarching arguments that necessarily homogenize different technologies. Betz and Cacean (2012) suggest that the moral arguments about CE can only be fully evaluated, when the placeholder T is substituted with a specific technology. Thus, placeholder T ought to be specified when evaluating the arguments. An overview of the macro-structure of the debate is seen in Figure 3.

Climate Engineering

Figure 3. Macro-structure of the debate on CE (Betz & Cacean, 2012, p. 21).

Restrictions on Research?

A central thesis within the argument map is the claim that CE should be researched. This claim is backed by three theses: (a) readiness for deployment is desirable; (b) side effects of research and development are negligible; and (c) there is no alternative to immediate research. All assumption may be, of course, questioned.

The strongest argument in favor of CE research is known as the arm the future argument (Betz, 2012). It starts with stating that there is uncertainty about the future states of mitigation and climate change, and especially a climate crisis cannot be ruled out. Facing uncertainty and looming evil, future generation should be bequeathed a full-fledged set of well-researched options. Via research, future generations are brought into a situation of responsible decision-making.

Reducing uncertainty on CE options seems mandatory, especially regarding the de facto desirability of any CE scheme. Research on ethical, political, and social aspects of a CE technology ought to be carried out in order to scrutinize the claim that its deployment can be, in fact, desirable. If, however, desirability or at least permissibility of CE deployment can be ruled out, then further research is uncalled-for. In other words, if deployment will be morally wrong or not wanted for other reasons, then why perform research at all? Consequently, the majority of the moral arguments about CE research presume the desirability or, at least, the permissibility of CE deployment under some emergency scenario, with the actual decision about deployment being left to future generations.

The most prominent argument against research is the so-called “moral hazard” argument (which can also be coined to attack the possible deployment of CE). It goes against the premise, that side effects of research will be negligible by pointing out that the prospect of actual CE deployment is implied in research. This prospect will diminish the efforts to reduce emissions (abatement). From the moment that CE deployment appears to be a real option, mitigation will seem unattractive. The moral hazard argument is a paradigm case of hypothetical reasoning. If, for instance, companies have to decide whether to invest billions either in renewable energies or in non-conventional fossil fuels (tar sand, deep sea drilling, methane hydrates), and if they get aware that large SAI research is underway, and if they believe in high likeliness of deployment, then they may decide in favor of fossil fuels. If companies have huge amount of capital in fossil fuel industries, they may also promote and support SAI. Large-scale research on SAI provides prospects for final deployment and the prospects as such may launch investments back into fossil fuels and, by doing so, prolong the current energy supply system. This line of reasoning could contribute to an argument from political economy of SAI (Ott, 2018). It is beyond the scope of this article to present all sophistications of the moral hazard debate. For ethical analysis of the moral hazard argument, see Baatz (2016), Hale (2009), Morrow (2014).

Another argument proposes to make large-scale research on SAI conditional (Baatz & Ott, 2017). The conditionality argument is based on the distinction between different types of research: model runs, experiments in laboratories, small-scale field experiments (Dykema, Keith, Anderson, & Weisenstein, 2014), and large field tests. The qualifier “large” means: on a scale sufficient enough to assess the efficacy of full SAI deployment on a global scale. Large field tests come close to partial deployment. The argument supposes that only states can finance such field tests by launching a big science research program. According to the argument, states are in a moral position to launch research programs which ramp up to large field tests in the face of the moral hazard if and only if they cut their emissions domestically and substantially, engage within the overall Paris process, and pay into adaptation finance schemes appropriately. Another condition may be prior and informed consent to such field tests being given by the general assembly of the United Nations. In the moral background of the conditionality argument are assumptions about trustworthiness. If an agent does not fulfill the above conditions but instead researches a technology which quickly reduces temperatures, this raises suspicions that the technology is actually supposed to be more like a replacement rather than a supplement. This undermines the consensus that CE should not replace mitigation. Such agents should be considered unsuitable to responsibly handle a risky technology such as SAI. This conditionality argument is based on supposition about trust and distrust in agents and states to launch “big science” research on SAI. Thus, it connects to the argument concerning moral corruption.

Most people involved in ethics of science and responsible research would favor a step-by-step approach, starting with modeling, contained experiments and small-scale experiments in open environments. Thus, small-scale experiments with ocean fertilization and SAI are permissible if the impacts on the environment remains low and brief (as in the case of LOHAFEX). But to show effects, experiments must be ramped up to large-scale field trials, as in the case of SAI. Such field tests are not justified by the freedom of research principle because they deeply interfere with planetary nature.

On Climate Engineering Deployment: Crucial Arguments

The notion to deploy any CE solution as mere replacement of mitigation is widely rejected. Neuber (2018) summarizes the discussion on CE and shows that no serious scholar proposes a CE-alone climate policy. CE technologies ought to be always embedded in a climate portfolio also including mitigation and adaptation, if deployed at all. However, a portfolio that exclusively relies on abatement and adaption (zero CE portfolio) has to face grave risks not to reach the objective to limit global warming well below 2°C. If so, some CE measures must be assets in a prudently chosen and risk averse portfolio.

Most arguments about CE concern the desirability or permissibility of its future deployment as a measure to combat the impacts of climate change.

Means-End Relationship

One argument in favor of CE deployment relates to the 2°C target. Mitigation efforts may not be enough to stabilize temperature at this level. There is a commitment to reach the goal. The goal can be reached by effective deployment. Hence, the deployment of CE seems mandatory. Well below 2°C is the goal; CE is the only means to reach the goal. Thus, it must be used. This argument relies on assumptions about the future development of international climate politics, which may be overly pessimistic. Assuming the necessity of CE because of insufficient mitigation efforts could also turn out to be a self-fulfilling prophecy and enhance a trade-off with mitigation. The means-end argument seems too straightforward. Imagine that without CE climate change would be in the magnitude of, say, 2.3°C in 2060 and would drop slowly afterwards. Imagine further that adaptation strategies work well and natural climate solutions may show effects in the longer run. Suppose that CE, especially SAI, are high-risk strategies. If so, it may be reasonable to reject SAI, even if it may be effective.

Costs of Deployment

Another line of reasoning pro potential CE deployment calls to the economic benefits of CE as part of a prudent climate portfolio in comparison to fast and far-reaching mitigation alone. CE could, for example, minimize the costs of energy transition, hence deploying it would be economically rational.

Those kinds of arguments are rebutted, for ease, and cost effectiveness are by no means the only issues when considering CE deployment. The argument focuses deployment costs only, and it ignores all opportunity costs, including external effects. Negative side effects of CE may seriously outweigh the benefits of quick and easy deployment. A more theoretical refusal argues that the idea to calculate an efficient (optimal) climate policy has lost credit because all calculation are based on normative assumptions (as rate of discount, value of a statistical life, monetization of ecosystem services). If so, it seems misleading to apply a generally flawed idea on the particular CE case. The argument that SRM may enable a less expensive transition of the global energy system is also based on many contestable assumptions on which economic models are based (growth rates, diffusion of technologies, returns of investments, carbon pricing schemes and more). Results of economic models should be taken as a piece of information only, not as decisive reasons for deploying SRM.

Moreover, Klepper and Rickels (2012) argue that the calculations of low SAI deployment costs are flawed. Perhaps, a strategy is faced, being familiar to many large-scale technological and architectural projects. In the initial stage of the project, low costs are presented by the proponents. If the project is going to be realized, costs increase, but the project continues nevertheless since a point of no return has been passed. Moriyama et al. (2016) present cost assessment that confirm Klepper and Rickels (2012). In any case, preparation period, aircraft construction, degree of intended cooling, altitude of injection, logistics, monitoring, and security are costly. Lockley (2019) provides a detailed outlook in the costly securitization equipment for SAI needed against different attack triggers, as cyberattacks, violent protesters, or even warfare. Security of SAI may give military forces a prominent role, since “the use of military airfields or dedicated geoengineering airbases offer a general step up in security” (Lockley, 2019, p. 10). Securitization must be included in deployment cost assessments, and it motivates general concern about the military dimension of SAI (see the section on “Dual Use”). The point about security may be of relevance for the political economy of SRM.

Lesser Evil

Among the strongest argument in favor of CE deployment is the “lesser evil” argument (Betz, 2012; Gardiner, 2010). The lesser evil argument assumes that there may be a future point in time where a choice between dangerous, unmitigated climate change, on the one hand, and the deployment of CE, especially fast-acting technologies like SAI, on the other, will have to be made. In this case, CE would clearly appear to be the lesser evil. The deployment of CE then would indeed be desirable (if not unavoidable), given the conditions. Because of this prospect, one should start research into CE technologies immediately, in order to have them ready in time and grant future decision-makers an additional option to their climate portfolio (arm the future). This argument can be refined by pointing to the victims of climate change. If, as many studies suggest, many adverse impacts of climate change will fall upon poor people in the Global South, there seems a humanitarian obligation to protect such people via SAI or other fast-acting CE schemes. This humanitarian line of reasoning has found some support in the literature (Buck, 2019). A variant of this argument says that it will be permissible for technologically well-equipped states to deploy SAI in their own national interest as long as the side effects of such interest-based deployment will benefit poor people in other world regions (Horton & Keith, 2016, p. 83).

The lesser evil line of reasoning, which is among the most prominent in favor of CE, can be attacked in various ways: for one, it is not clear that CE will really be the lesser evil. Taking into account the many unclear side effects, it is debatable what is seen as less or evil (Betz, 2012). It is by no means obvious that the deployment of CE will be the lesser evil as compared to (unmitigated) climate change.

Additionally, as Gardiner (2010) argues, there are choices that may be tarnished in any case. In a lesser evil scenario, decision-makers would have to choose between a catastrophic climate change on the one side, and possibly dangerous climate manipulation on the other. In this dilemmatic situation, there are no right choices—in any case, the innocent will suffer. Gardiner argues that any situation in which this tarnished choice arises ought to be avoided at all costs. So, Gardiner does not simply attack the weighting mechanism that lies within the lesser evil argument, but rather, points out that the lesser evil argument only arises if a morally unacceptable situation is taken for granted. To Gardiner, our primary obligation is to prevent dilemmatic situations, but not to react upon such situations or, even worse, force other people to decide within such situations.

While Gardiner’s reasoning is very compelling, it must assume the possibility to avert the lesser evil situation. But the later global decision-makers tighten the climate goals, the less likely stabilizing at well below 2°C becomes. And the better the mechanism of fast-acting CE are understood, the more probable the lesser evil situation may become, if there is a moral hazard. Hence, the lesser evil argument may turn out to be a self-fulfilling prophecy, where the preparation for such dire scenario may bring it about. This can be called not just the dilemma, but rather the tragedy of the lesser evil framing.

The very premise of the lesser evil argument may be rejected as a deceptive scenario. It is unlikely that there will be unmitigated climate change. Mitigation policies are underway and taking momentum in many regions, branches of industries, companies, township communities, and individual agents. There is divestment away from fossil fuel industries. Following the lesser evil scenario ignores the many efforts to mitigate. Unmitigated climate change would only be likely if the fossil fuel empire may strike back successfully and international climate policies collapse.

Emergency Case

The lesser evil argument is akin to the emergency argument. In a climate emergency, where global climate impacts happen fast and intense, CE may serve as a back-up plan, an insurance that serves as a shield against this kind of catastrophe (Betz, 2012; Hamilton, 2013; Keith, 2013; Uther, 2013). Lately, the emergency framing of climate change has been challenged. Invoking an emergency situation in order to justify the deployment of a risk technology may lead to problematic political and social implications (Horton, 2015). Sillmann et al. (2015) argue that the emergency argument fails for scientific and political reasons. If tipping points have been touched and thresholds have been crossed, SAI will not be able to stop and reverse the consequences. Sillmann et al.(2015, p. 291) argue that the potential for SRM “to respond effectively to tipping-point emergencies is very restricted”. The emergency argument is not based on scientific findings on tipping points (Lenton et al., 2008).

From a political perspective, it is not clear what would count as an (global) emergency situation and who has the power to define it. If an emergency state was to be declared, power may concentrate in the hands of a few, with the associated risk of abuse (Horton, 2015). Remember Carl Schmitt: “Sovereign is he who decides on the exception” (Schmitt 2006, p. 36). The legal status and political background of an emergency situation may threaten deliberative, democratic structures. Additionally, as Gardiner (2010) has prominently pointed out, there is no moral obligation to prepare for an emergency situation as long as the capacity to avoid it (or at least reduce it) is still at hand.

Buying Time

The rebuttal of the emergency framing also gives more attractiveness to a very prominent argument: the buying time argument. Instead of preparing for an emergency scenario that could arguably still be avoided, CE could be used as a stopgap measure to buy time until abatement polices show effect on a global scale.

The buying time strategy has gained popularity since the 2010s in academic discourse. It represents a seemingly prudent way to benefit from CE deployment while minimizing its negative side effects. Starting from the assumption that climate change will show dire effects in the second half of the 21st century, but emissions will reach net zero earliest at the end of the century, some authors propose CE as a stopgap measure for this interim period (MacMartin, Ricke, & Keith, 2018). The discrepancy in time can be bridged by the use of some CE strategy which will cushion the effects of climate change, while buying time for effective mitigation measures. It is part of the debate whether all CE technologies can act as a time buying bridge technology or whether the BTA only applies to SRM technologies. Since CDR takes time to show effect, SRM may be the better candidate for a stopgap measure. However, as shown in Neuber (2018), long-term strategies including the deployment of BECCS can count as buying time strategies. Thus, the argument is clearly technology and portfolio-specific.

BTA is ridden with prerequisites. To start with, the BTA assumes that the decline in global GHG-emissions correlates with a decline in climate change impacts. Due to CO2 inertia, this is not the case (Solomon et al., 2009). So the question really is: buying time for what? The answer to this question is to buy more time for an effective climate policy mix that will eventually guarantee the well-below 2°C target. This policy mix ought to ensure a non-catastrophic climate change by itself, making CE an add-on that is deployed as needed and ramped down as quickly as possible. This is a rather strong normative framing for the use of CE. This normative frame is expressed by the following five preconditions to the BTA:

  1. 1. The use of CE is time-limited and will be ceased as soon as its goal is reached.

  2. 2. Aggressive mitigation strategies are undertaken in parallel.

  3. 3. The use of CE does not lead to a decline in mitigation efforts.

  4. 4. The use of CE is shaped to have no or very little negative side effects.

  5. 5. The use of CE is not morally forbidden.

Those five conditions are each necessary and together they are sufficient to make CE deployment morally sound. If they can be affirmed with high confidence, CE deployment can be permissible, acceptable, or even desirable. If the BTA, then, renders CE even obligatory is left open for further discussion. These contested points indicate that the BTA is anything but an easily achieved common ground in the debate on CE. Rather, it is highly demanding for policymakers and scientists alike, to guarantee the fulfillment of the five prerequisites. If the BTA is to work, it should not leave room for empty promises (Neuber, 2018).

Environmental and Societal Impacts

The largest group of arguments against CE deployment consequently relates to potential side effects of deployment. Critics of CE fear severe and uncontrollable effects on the environment (Robock, 2008). Hence, they oppose the deployment and development of CE technologies. For advocates of research precisely this uncertainty is reason to deepen the research. Environmental concern refers to precipitation patterns, the ongoing of ocean acidification, and influences on plant growth. Some model studies predict an enhancement of the terrestrial photosynthesis rate (Xia, Robock, Times, & Neely, 2016), while others emphasize uncertainties of the vegetation response to SAI (Glienke, Irvine, & Lawrence, 2016). Despite many model studies, the environmental impacts of a combination of CDR and SRM remain uncertain to a large extent. SAI could cause acid rain because sulfate aerosols will drop down to Earth after some time. Of course, the magnitude of this effect may be little compared to other anthropogenic sources of acid rain. In absolute quantities, it is substantial. Here, again, robust and trustworthy simulations are due in order to minimize environmental hazards. The effects of SAI on ocean acidification is at best unclear (Böhm & Ott, 2019; Keller, Feng, & Oschlies, 2014). SAI may affect ocean circulation and precipitation patterns, and it may delay the recovery of the ozone layer. In any case, environmental risks of SAI should not be downplayed.

Additionally, the deployment of CE may lead to great injustice. Especially with SAI, studies indicate that the ecological impacts will be distributed unequally, potentially harming the world’s vulnerable more than benefit them. Especially native peoples that contribute the least to global warming may be affected substantially by artificial climate modification. The question of consent becomes essential in such cases (Whyte, 2012). It is fair to note that indigenous people do not agree on the “humanitarian” argument that SAI may help the global poor.


Taking the numerous, but uncertain side effects into account, CE has been criticized as a form of human hubris that aims at controlling nature regardless the uncertain consequences. Such hubris arguments claim that CE shows the failure of a people that cannot live on a given environment without destroying it (“fouling the nest”; Fleming, 2010; Gardiner, 2012). The hubris argument, however, has a metaphysical smell. In Greek antiquity, it points to overbearance of humans. In Christianity, it became a “playing God” argument. Without monotheistic beliefs, playing God becomes a metaphor that must be made explicit on secular grounds. In a secular age, hubris arguments may, at best, point at the phenomenon that scientists and engineers often overrate their capacity to regulate and control CE schemes on, a long-term planetary scale. Overrating one’s technological, political, and moral capabilities may be true a fortiori for SAI. Such hubris may conjoin with moral corruption in a specific attitude of playful tinkering with the global thermostat. Ultimately, this line of reasoning must argue in favor of humility with respect to the global climate system and the biosphere. An anthropological reflection upon the hubris argument from a theological perspective is given by Clingerman (2014). Finally, Clingerman (2014, p. 18) seems to affirms CE since the world has to accept “that someone controls the Earth’s thermostat.”

Political Economy

Other arguments against SAI are arguments about the political economy of SAI (Gunderson, Petersen, & Stuart, 2018; Long & Scott, 2013; Ott, 2018; Sikka, 2012). It is argued that some agency networks in specific countries may propagate SAI out of vested interests. These PE arguments, however, rest on many contested premises. In a nutshell, they suppose a specific susceptibility of specific economic and political system to research and, finally, deploy SAI to secure business as usual being based on fossil fuels. If such structures become dominant in a given order, influential agency networks propose and campaign for SAI at the interfaces of science, policy boards, think tanks, mass media, and political administration. Proponents will design a specific pro-SAI framing, including moral narratives, efficiency calculations, and national security against climate emergencies. The vested interest behind pro-SAI rhetoric may be the representatives of capitals being fixed in fossil fuel industries because such capital will lose economic value under a global de-carbonization trajectory. PE arguments have been applied to the United States. Those specific analyses rely on analytic Marxism, but are not committed to any leftist policy proposals. PE arguments can be combined with the conditionality argument against large SAI research.

Technology-Specific Arguments

While there are evaluation patterns that apply to CE in general or to some technology groups, there may be nevertheless single features of CE assets that are obscured by generalization. The placeholder T in the argument map must be specified in order to avoid lofty talk on CE. This is true not only of the technologies themselves, but also of the portfolios in which they are embedded. Each CE technology should be evaluated in a comprehensive climate portfolio, where both the extent of its planned deployment as well as the additional climate options like mitigation is taken into account. To illustrate this, there will be a brief discussion on some arguments which are not applicable to all CE technologies, or, even if they are technology-specific, do not apply to all possible deployment scenarios.

Emergency Argument

Some authors see CE as a kind of emergency insurance, in case climate change is faster and more severe than expected. CE could then help to stabilize temperature and to avert the worst effects. The insurance metaphor is closely linked to the emergency argument. While all efforts should be put into preventing a climate catastrophe, the damage could be so immense that preparation for its occurrence should be under way. An insurance measure should quickly and effectively balance out the catastrophe against which it is made, and compensate for its damage. Hence, the focus of authors adopting this emergency metaphor lies on the controllability and economic rationality of CE measures (Luokkanen, Huttunen, & Hildén, 2013). This may be the case especially for SAI, since it is deemed to be easy to apply, presumably cheap in its deployment and maintenance costs and rather effective in cooling the Earth. The emergency argument only holds for SAI. SAI may be seen as an insurance means of last resort in case of emergency. The criticism against the emergency argument flies in the face of the insurance argument.

The emergency framing loses its impact when used for CDR measures. Due to the physical inertia of the atmosphere, effectiveness of CDR is only visible after years or even decades. While during an abrupt temperature rise with potentially irreversible effects, SAI could be used quickly and effectively to stabilize temperature artificially, CDR can only stabilize temperature in the long term. Emergency considerations do not apply to CDR technologies.

Termination Problem

If SAI is used as a substitute for aggressive abatement, atmospheric CO2 concentration will continue to rise. Only the effect of high CO2 concentration, namely the temperature rise, will be artificially stabilized. An abrupt termination of SAI (for example, due to social and political upheavals) would lead to an accelerated heating enforced by the large volume of atmospheric CO2 (Robock, 2008; Royal Society, 2009). This may be disastrous for the climate and the biosphere. Additionally, this situation could put future generations in a dilemmatic situation: either they would have to decide to continue to operate SAI, with possibly fatal side effects, or they would have to accept accelerated climate change (Ott, 2012).

The termination problem is obviously coined for a specific case: SAI deployment without adequate abatement. It is easy to see that SAI does not lead to a termination problem when it is accompanied with drastic emission cuts—for example, in a buying time scenario. The argument only holds for an abrupt stop of SAI. It must be refined if termination of SAI is performed gradual over a longer period in time in combination with abatement. Nevertheless, the termination problem still gives reasons to be concerned. It presupposes a somewhat stable political governance capacity to deploy, control, secure, assess, and terminate SAI. Abrupt termination due to political crisis remains possible. The termination problem connects to the hubris argument: the capacities may be overrated to govern and control SAI. According to Parker and Irvine (2018), however, a geographically dispersed SAI deployment hardware being maintained by a coalition of powerful states could safeguard against most kinds of drivers causing involuntary termination (as terrorism). Interesting, They mention damaging environmental side effects as presumptive reasons to terminate SAI in the outline of their article, but only address (a) destruction of deployment infrastructure; (b) catastrophe; and (c) elective termination. As they rightly point out, elective political efforts for termination may cause high-stake conflicts with powers wishing to maintain SAI. Clearly, such scenarios are speculative.

The termination problem does not apply to any types of CDR measures—for they do address the root cause of climate change by reducing the CO2 content in the atmosphere. Even if CDR measures were used as an alternative to and a substitute for CO2 emission reduction, and were to be discontinued, there would be no rapid temperature rise as in the illustrated SAI case.

Dual Use

A very special problem is connected to SRM technologies. Due to their influencing the radiative balance, they can manipulate local weather patterns with potential negative side effects. For example, local floods or droughts can be triggered by SAI. This makes SAI and other SRM technologies potential weapons of warfare (Corner & Pidgeon, 2010; Goodell, 2010). In fact, the idea of weather manipulation as an instrument of warfare goes back a least to the middle of the 20th century, and much of the research on SAI and other SRM technologies is based on military research of similar schemes (Fleming, 2010). Weather modification for the purpose of warfare is prohibited by international convention, and the rearmament of these technologies could already be read as a violation of the convention (Corner & Pidgeon, 2010).

Securitization of SAI may also contribute to militarization. Lockley (2019) considers anti-missile missile systems, gun systems, air-to-air-missiles in SAI aircrafts, and aircraft carriers being protected by submarines. Obviously, however, this military risk is a speculative worst-case scenario for SRM or SAI and, perhaps, even for marine cloud brightening which could have regional impacts. It is unlikely to conceive a weapon of mass destruction being derived from long-term CDR measures or from NCS. Some specific forms of CDR, like ocean fertilization, may, however, be used as a form of large-scale pollution with hostile intentions. Irrigated BECCS on a large scale may intensify political upstream or downstream conflicts over freshwater. Such scenarios, however speculative, should spur debates on the regulation of CE by means of international law and governance schemes.

Closing Remarks

This article has tried to give an overview on both some CE technologies and the moral issues surrounding research and/or possible deployment of CE. But it may fall short to discuss CE in an abstract generic mode, since it comprises technologies with very different risk profiles and moral issues. Even more so, every single deployment scheme ought to be judged specifically, for it is a huge difference to propose SAI, for example, as a substitute for abatement, or to embed it within a comprehensive climate portfolio including abatement and adaptation, where SAI will be used sporadically and only for a matter of decades.

If the central moral arguments pro and contra CE are conjoined with the most prominent deployment schemes, a discursive matrix will arise:

Technology at stake Deployment scenario 1

Technology at stake Deployment scenario 2

Technology at stake Deployment scenario 3

Argument A

Argument B

Argument C

The evaluation of the single entries of this matrix helps to understand the vast landscape of CE and allows for robust decisions about the practicality as well as the moral soundness of certain CE deployment schemes to be made. To repeat the methodical precept:

  1. 1. Specify the CE technology at stake.

  2. 2. Conceive the technology as an asset in the portfolio (costs, potential).

  3. 3. Conceive a comprehensive risk assessment.

  4. 4. Take the difference between research, testing, and deployment into account.

  5. 5. Consider presumptive pro and con arguments being embedded in the maps.

  6. 6. Make your argument!

  7. 7. Reflect upon the (meta)ethical suppositions in your argument.

  8. 8. Assess implications for portfolio strategies and governance schemes.

Common Moral Ground: Climate Engineering as Complementary Strategy

Between the support of large-scale CE as a strategy to realize climate goals (well below 2°C) in the face of insufficient mitigation on the one hand, and the moral devaluation of CE on the other, an apparent middle ground of reasonable agreement has been sought. There is some commonly shared moral ground:

  1. 1. CE deployment should not amount to an alternative to abatement efforts.

  2. 2. CE should not be offered as a simple and cheap solution to the broader public and to policymakers (Corner & Pidgeon, 2010; Hale, 2012; Lin, 2013; Morrow, 2014).

  3. 3. Strategies that render SAI unnecessary, deserve special attention.

If there is any consensus within the CE debate, it is the consensus that CE is a complement (supplement) to abatement, not an alternative or substitute. Even advocates of SAI research point out that CE and emission reduction are no alternatives, but rather equal parts of a comprehensive climate policy portfolio. If so, a simple economics of substitution should not be applied to the relation between CE and abatement (see Reynolds, 2014b). Many arguments in favor of CE assume or demand that CE deployment will be accompanied by serious mitigation efforts. Under this perspective, the assessment of CE depends on the supporting climate strategies like abatement and adaptation. The need to balance the portfolio in favor of abatement is emphasized in Bahn, Chesney, Gheyssens, Knutti, and Pana (2015). Even proponents of SAI reject the idea that SAI may be deployed as a cheap and easy techno-fix that makes emission cuts unnecessary. It seems worth noting that such crude techno-fix hopes are no longer part of a serious CE debate. The idea to combine high GHG concentrations in the lower atmosphere with a sulfate-based protective layer in the stratosphere, which must be hold in balance over centuries, does not find any support within the CE discourse. There is no reasonable portfolio strategy without aggressive abatement.

Rather, CE presents itself as a presumptive stopgap measure that allows for more time, while in the interim period an ambitious emission reduction program will be underway in order to push the decarbonization of society forward. The deployment of CE, including SRM, could help to buy time for initiating a successful decarbonization policy in the longer run. This way, CE may act as a bridge technology that could be ramped down as soon as mitigation efforts show sufficient effect. This line of thinking amounts to the buying time framing of climate engineering (Neuber, 2018). Therefore, the buying time argument deserves special attentiveness and close investigation in a special article.

Realistic Utopia

Steffen et al. (2018) have argued that collective human action may, despite global warming and positive feedbacks, avoid uncontrollable hothouse Earth and stabilize the Earth system in a habitable interglacial state. The task is to push back the Earth system from the trajectory that may collapse in a basin of attraction toward hothouse Earth. From the perspective of climate ethics (Ott, 2011), environmental ethics, and a theory of strong sustainability (Ott, 2014), the most attractive pathway would be the following one:

  1. 1. Fast and aggressive abatement of emissions on the global scale. Emissions should peak globally within a few years (2030) and sink afterwards as soon as possible. The national contributions should become more ambitious over time. Single countries should take forerunner roles, as in phasing out coal.

  2. 2. Adaptation finance should be generous (100 billions per year), but limited. There should be multiple criteria (beyond vulnerability) to determine how to launch the money into desirable projects. One crucial criteria should be to natural climate solutions and strong sustainability. This would bring connectivities between UNFCCC and CBD.

  3. 3. Increasing the price for carbon either by tax or by cap-and-trade systems should mobilize money to finance NCS on larger scales.

  4. 4. Research on NCS and priority for funding are intensified.

  5. 5. A change to more vegetarian diets releases pressure on land and enhances prospects for BECCS. Developed countries should start BECCS within their territory even if there may be some protests against dumping CO2 underground.

  6. 6. The only credible SAI strategy is based on the buying time argument and to be seen as means of the last resort. The risks of SAI must be carefully weighed against a temporal overshoot (+2.4°C GMT). Deployment of SAI requires the global prior informed consent within the system of the United Nations.


This article is an outcome of the Priority Program 1689 of the German Research Foundation DFG “Climate Engineering: Risks, Challenges, Opportunities?”


Adger, N., Lorenzoni, I., & O’Brien, K. (Eds.). (2009). Adapting to climate change. Cambridge, UK: Cambridge University Press.Find this resource:

Anderson, C. M., DeFries, R. S., Litterman, R., Matson, P. A., Nepstad, D. C., Pacala, S., … Field, C. B. (2019). Natural climate solutions are not enough. Science, 363, 933–934.Find this resource:

Anderson, K., & Peters, G. (2016). The trouble with negative emissions. Science, 354, 182–183.Find this resource:

Armeni, C., & Redgwell, C. (2015). International legal and regulatory issues of climate geoengineering governance: Rethinking the approach. Climate Geoengineering Governance Working Paper Series 021.Find this resource:

Baatz, C. (2016). Can we have it both ways? On potential trade-offs between mitigation and solar radiation management. Environmental Values, 25(1), 29–50.Find this resource:

Baatz, C. (2018). Climate adaptation finance and justice. Analyse & Kritik, 40(1), 73–105.Find this resource:

Baatz, C., & Ott, K. (2017). Why aggressive mitigation must be part of any pathway to climate justice. In C. Preston (Ed.), Climate justice and geoengineering (pp. 93–108). London, UK: Rowman & Littlefield.Find this resource:

Bahn, O., Chesney, M., Gheyssens, J., Knutti, R., & Pana, A. (2015). Is there room for geoengineering in the optimal climate policy mix? Environmental Science & Policy, 48, 67–76.Find this resource:

Betz, G. (2012). The case for climate engineering research: An analysis of the “arm the future” argument. Climatic Change, 111, 473–485.Find this resource:

Betz, G., & Cacean, S. (2012). Ethical aspects of climate engineering. Karlsruhe, Germany: KIT Scientific Publishing.Find this resource:

Böhm, F., & Ott, K. (2019). Impacts of ocean acidification. Marburg, Germany: Metropolis.Find this resource:

Bonsch, M., Humpenöder, F., Popp, A., Bodirsky, B., Dietrich, J. P., Rolinski, S., …, Stevanovic, M. (2016). Trade-offs between land and water requirements for large-scale bioenergy production. Global Change Biology Bioenergy, 8(1), 11–24.Find this resource:

Boucher, O., Forster, P., Gruber, N., Ha-Duong, M., Lawrence, M., Lenton, T., …, Vaughan, N. (2014). Rethinking climate engineering categorization in the context of climate change mitigation and adaptation. WIREs Climate Change, 5(1), 23–35.Find this resource:

Buck, H. (2019). After geoengineering: Climate tragedy, repair, and restoration. London, UK: Verso.Find this resource:

Clingerman, F. (2014). Geoengineering, theology, and the meaning of being human. Zygon, 49(1), 6–21.Find this resource:

Corner, A., & Pidgeon, N. (2010). Geoengineering the climate: The social and ethical implications. Environment, 52, 24–37.Find this resource:

Creutzig, F. (2014). Economic and ecological views on climate change mitigation with bioenergy and negative emissions. Global Change Biology Bioenergy, 8(1), 4–10.Find this resource:

Crutzen, P. (2006). Albedo enhancement by stratospheric sulfur injections: A contribution to resolve a policy dilemma? Climatic Change, 77, 211–220.Find this resource:

Dykema, J., Keith, D., Anderson, J., & Weisenstein, D. (2014). Stratospheric-controlled perturbation experiment: A small-scale experiment to improve understanding of the risks of solar geoengineering. Philosophical Transactions of the Royal Society A, 372, 20140059.Find this resource:

Fleming, J. R. (2010). Fixing the sky: Checkered history of weather and climate control. New York, NY: Columbia University Press.Find this resource:

Gardiner, S. (2010). Is “arming the future” with geoengineering really the lesser evil? Some doubts about the ethics of intentionally manipulating the climate system. In S. Gardiner, S. Caney, D. Jamieson, & H. Shue (Eds.), Climate ethics: Essential readings (pp. 284–312). Oxford, UK: Oxford University Press.Find this resource:

Gardiner, S. (2012). Are we the scum of the earth? Climate change, geoengineering and humanity’s challenge. In A. Thompson & J. Bendik-Keymer (Eds.), Ethical adaptation to climate change. Human virtues of the future (pp. 241–260). Cambridge, MA: MIT Press.Find this resource:

Gardiner, S. (2013). Geoengineering and moral schizophrenia: What is the question? In W. Burns & A. Strauss (Eds.), Climate change geoengineering: Legal, political and philosophical perspectives (pp. 11–38). Cambridge, UK: Cambridge University Press.Find this resource:

Gasparini, B., & Lohmann, U. (2016). Why cirrus cloud seeding cannot substantially cool the planet. Journal of Geophysical Research Atmospheres, 121(9), 4877–4893.Find this resource:

Glienke, S., Irvine, P., & Lawrence, M. (2016). The impact of geoengineering on vegetation in experiment G1 of the GeoMIP. Journal of Geophysical Research Atmospheres, 120(10), 196–213.Find this resource:

Goodell, J. (2010). How to cool the Planet: Geoengineering and the Audacious Quest to fix Earth’s climate. New York: Houghton Mifflin.Find this resource:

Griscom, B. W., Adams, J., Ellis, P. W., Houghton, R. A., Lomax, G., Miteva, D. A., …, Fargione, J. (2017). Natural climate solutions. Proceedings of the National Academy of Sciences, 114(44), 11645–11650.Find this resource:

Gunderson, R., Petersen, B., & Stuart, D. (2018). A critical examination of geoengineering: Economic and technological rationality in social context. Sustainability, 10(1), 269.Find this resource:

Hale, B. (2009). What’s so moral about the moral hazard? Public Affairs Quarterly, 23(1), 1–25.Find this resource:

Hale, B. (2012). The world that would have been. Moral hazard arguments against geoengineering. In C. Preston (Ed.), Engineering the climate: The ethics of solar radiation management (pp. 113–132). Lanham, MD: Rowman & Littlefield.Find this resource:

Hamilton, C. (2013). Earthmasters: The dawn of the age of climate engineering. New Haven, CT: Yale University Press.Find this resource:

Hansen, J., Sato, M., Hearty, P., Ruedy, R., Kelley, M., Masson-Delmotte, V., …, Lo, K.-W. (2016). Ice melt, sea level rise and superstorms: Evidence from paleoclimate data, climate modeling and modern observations that 2°C global warming could be dangerous. Atmospheric Chemistry and Physics, 16, 3761–3812.Find this resource:

Heyward, C. (2013). Situating and abandoning geoengineering: A typology of five responses to dangerous climate change. Political Science and Politics, 46(1), 23–27.Find this resource:

Horton, J. (2015). The emergency framing of solar geoengineering: Time for a different approach. The Anthropocene Review, 2(2), 147–151.Find this resource:

Horton, J., & Keith, D. (2016). Solar engineering and obligations to the global poor. In C. Preston (Ed.), Climate justice and geoengineering. London, UK: Rowman & Littlefield, pp. 79–92.Find this resource:

IPCC (International Panel on Climate Change). (2007). Climate change 2007: Synthesis report. contribution of working groups I, II and III to the fourth assessment report of the intergovernmental panel on climate change [Core Writing Team, R. K., Pachauri, & A. Reisinger (Eds.)]. Cambridge, UK: Cambridge University Press.Find this resource:

IPCC (International Panel on Climate Change). (2013). Summary for policymakers. In T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, …, P. M. Midgley (Eds.), Climate change 2013: The physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change (pp. 3–29). Cambridge, UK: Cambridge University Press.Find this resource:

IPCC (International Panel on Climate Change). (2014a). Summary for policymakers. In O. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, …, J. C. Minx (Eds.), Climate change 2014: Mitigation of climate change. Contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change (pp. 1–30). Cambridge, UK: Cambridge University Press.Find this resource:

IPCC (International Panel on Climate Change). (2014b). Summary for policymakers. In C. B., Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir, …, L. L.White (Eds.), Climate change 2014: Impacts, adaptation, and vulnerability. Part A: Global and sectoral aspects. Contribution of working group II to the fifth assessment report of the intergovernmental panel on climate change (pp. 1–32). Cambridge, UK: Cambridge University Press.Find this resource:

IPCC (International Panel on Climate Change). (2014c). Climate change 2014: Synthesis report. Contribution of working groups I, II and III to the fifth assessment report of the intergovernmental panel on climate change [Core Writing Team, R. K. Pachauri, & L. A. Meyer (Eds.)]. Geneva, Switzerland: IPCC.Find this resource:

Jacquet, J., & Jamieson, D. (2016). Soft but significant power in the Paris Agreement. Nature Climate Change, 6, 643–646.Find this resource:

Keith, D. (2013). A case for climate engineering. Boston, MA: Boston Review Books.Find this resource:

Keller, D., Feng, E. Y., & Oschlies, A. (2014). Potential climate engineering effectiveness and side effects during a high carbon dioxide-emission scenario. Nature Communications, 5, 3304.Find this resource:

Klepper, G., & Rickels, W. (2012). The real economics of climate engineering. Economics Research International, 2012, 1–20.Find this resource:

Lenton, T., Held, H., Kriegler, E., Hall, J., Lucht, W., Rahmstorf, S., & Schellnhuber, H. J. (2008). Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Sciences, 105(6), 1786–1793.Find this resource:

Lewis, S. L., Wheeler, C. E., Mitchard, E. T., & Koch, A. (2019). Regenerate natural forests to store carbon. Nature, 568, 25–28.Find this resource:

Lin, A. (2013). Does geoengineering present a moral hazard? Ecology Law Quarterly, 40, 673–712.Find this resource:

Lockley, A. (2019). Security of solar radiation management geoengineering. Frontiers of Engineering Management, 6(1), 102–116.Find this resource:

Lohmann, U., & Gasparini, B. (2017). A cirrus cloud climate deal? Science, 357(6348), 248–249.Find this resource:

Long, J. (2016). Bringing geoengineering into the mix of climate change tools. In C. Preston (Ed.), Climate justice and geoengineering. Ethics and policy in the atmospheric anthropocene (pp. 109–120). London, UK: Rowman & Littlefield.Find this resource:

Long, J. C. S., & Scott, D. (2013). Vested interests and geoengineering research. Issues in Science and Technology, 29(3), 45–52.Find this resource:

Luokkanen, M., Huttunen, S., & Hildén, M. (2013). Geoengineering, newsmedia and metaphors: Framing the controversial. Public Understanding of Science, 23(8), 966–981.Find this resource:

MacMartin, D. G., Ricke, K. L., & Keith D. W. (2018). Solar geoengineering as part of an overall strategy for meeting the 1.5◦C Paris target. Philosophical Transactions of the Royal Society, 376, 20160454.Find this resource:

McNutt, M., Abdalati, W., Caldeira, K., Doney, S., Falkowski, P. G., Fetter, S. . . . Wilcox, J. (2015a). Climate intervention: Reflecting sunlight to cool earth. Washington, DC: The National Academies of Sciences, Engineering, and Medicine.Find this resource:

McNutt, M., Abdalati, W., Caldeira, K., Doney, S., Falkowski, P. G., Fetter, S. . . . Wilcox, J. (2015b). Climate intervention: Carbon dioxide removal and reliable sequestration. Washington, DC: The National Academies of Sciences, Engineering, and Medicine.Find this resource:

Moriyama, R., Sugiyama, M., Kurosawa, A. I, Masuda, K., Tsuzuki, K., & Ishimoto, Y. (2016). The cost of stratospheric climate engineering revisited. Mitigation and Adaptation Strategies for Global Change, 22(8), 1207–1228.Find this resource:

Morrow, D. (2014). Ethical aspects of the mitigation obstruction argument against climate engineering research. Philosophical Transactions of the Royal Society A, 372.(2031).Find this resource:

Niemeier, U., & Tilmes, S. (2017). Sulfur injections for a cooler planet. Science, 357, 246–248.Find this resource:

Niemeier, U., Schmidt, H., & Timmreck, C. (2011). The dependency of geoengineered sulfate aerosol on the emission strategy. Atmospheric Science Letters, 12, 189–194.Find this resource:

Neuber, F. (2018). Buying time with climate Engineering? An analysis of the buying time framing in favor of climate Engineering. DISSERTATION. KIT scientific publishing.Find this resource:

Ott, K. (2011). Domains of climate ethics. Jahrbuch für Wissenschaft und Ethik, 16, 95–114.Find this resource:

Ott, K. (2012). May solar radiation management constitute a dilemma? In C. Preston (Ed.). Engineering the climate (pp. 33–42). Lanham, MD: Lexington.Find this resource:

Ott, K. (2014). Institutionalizing strong sustainability: A Rawlsian perspective. Sustainability, 6, 894–912.Find this resource:

Ott, K. (2017). Normative pragmatics: Approach, promise, outlook. In J. P. Brune, R. Stern, Robert, & M. Werner (Eds.), Transcendental arguments in moral theory (pp. 213–229). Berlin, Germany: DeGruyter.Find this resource:

Ott, K. (2018). On the political economy of solar radiation management. Frontiers in Environmental Sciences, 6, 43.Find this resource:

Parker, A., & Irvine, P. (2018). The risk of termination shock from solar geoengineering. Earth’s Future, 6(3), 456–467.Find this resource:

Pelling, M. (2011). Adaptation to climate change. London, UK: Routledge.Find this resource:

Preston, C. (2013). Ethics and geoengineering: Reviewing the moral issues raised by solar radiation management and carbon dioxide removal. Wiley Interdisciplinary Reviews: Climate Change, 4, 23–37.Find this resource:

Rayner, S., Heyward, S., Kruger, T., Pidgeon, N., Redgwell, C., & Savulescu, J. (2013). The Oxford principles. Climatic Change, 121(3), 499–512.Find this resource:

Rescher, N. (1964). Hypothetical reasoning. Amsterdam, The Netherlands: North-Holland Publishing.Find this resource:

Reynolds, J. (2014a). The international regulation of climate engineering: Lessons from nuclear power. Journal of Environmental Law, 26(2), 269–289.Find this resource:

Reynolds, J. (2014b). A critical examination of the climate engineering moral hazard and risk compensation concern. The Anthropocene Review, 2(2), 174–191.Find this resource:

Reynolds, J. (2015). The international legal framework for climate engineering. Geoengineering our climate? Ethics, politics and governance, Working paper Series.Find this resource:

Reynolds, J., Parker, A., & Irvine, P. (2016). Five solar geoengineering tropes that have outstayed their welcome. Earth’s Future, 4(12), 562–568.Find this resource:

Rickels, W., Klepper, G., Dovern, J., Betz, G., Brachatzek, N., Cacean, S., …, Zürn, M. (2011). Large-scale intentional interventions into the climate system? Assessing the climate engineering debate. Kiel, Germany: Kiel Earth Institute.Find this resource:

Robock, A. (2008). Twenty reasons why geoengineering may be a bad idea. Bulletin of the Atomic Scientists, 64, 14–18.Find this resource:

Royal Society. (2009). Geoengineering the climate. Science, governance and uncertainty. London: Royal Society.Find this resource:

Schmitt, C. (2006). Political theology. University of Chicago Press. (Original: Politische Theologie, 1922).Find this resource:

Sikka, T. (2012). A critical discourse analysis of geoengineering advocacy. Critical Discourse Studies, 9(2), 163–175.Find this resource:

Sillmann, J., Lenton, T. M., Levermann, A., Ott, K., Hulme, M., Benduhn, F., & Horton, J. B. (2015). Climate emergencies do not justify engineering the climate. Nature Climate Change, 5, 290–292.Find this resource:

Solomon, S., Plattner, G.-K., Knutti, R., & Friedlingstein, P. (2009). Irreversible climate change due to carbon dioxide emissions. Proceedings of the National Academy of Sciences of the United States of America, 106, 1704–1709.Find this resource:

Steffen, W., Rockström, J., Richardson, K., Lenton, T., Folke, C., Liverman, D., … Schellnhuber, H. J. (2018). Trajectories of the Earth system in the anthropocene. Proceedings of the National Academy of Sciences, 115(33), 8252–8259; first published August 6, 2018.Find this resource:

Teller, E., Hyde, R., & Wood, L. (2002). Active climate stabilization: Practical physics-based approaches to prevention of climate change. Report Paper UCRL-JC-148012. Lawrence Livermore National Laboratory.Find this resource:

UNFCCC (Unites Nations Framework Convention on Climate Change). (2016). Adoption of the Paris Agreement. Paper presented at the Conference of the parties, Twenty-first session Paris, November 30 to December 11, 2015.Find this resource:

Uther, S. (2013). Diskurse des Climate Engineering—Argumente, Akteure und Koalitionen in Deutschland und Großbritannien. Heidelberg, Germany: Springer.Find this resource:

van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., …, Rose, S. K. (2011). The representative concentration pathways: An overview. Climatic Change, 109, 5–31.Find this resource:

Victor, D. (2008). On the regulation of geoengineering. Oxford Review of Economic Policy, 24(2), 322–336.Find this resource:

Victor, D., Morgan, G., Apt, J., Steinbruner, J., & Ricke, K. (2009). The geoengineering option. A last resort against global warming? Foreign Affairs, 88, 64–76.Find this resource:

Wagner, G., & Zeckhauser, R. J. (2012). Climate policy: Hard problem, soft thinking. Climatic Change, 110(3), 507–521.Find this resource:

Whyte, K. (2012). Indigenous peoples, solar radiation management and consent. In C. Preston (Ed.), Engineering the climate: The ethics of solar radiation management (pp. 65–76). Lanham, MD: Rowman & Littlefield.Find this resource:

Xia,. L., Robock, A., Times, S., & Neely III, R. (2016). Stratospheric sulfate geoengineering could enhance the terrestrial photosynthesis rate. Atmospheric Chemistry and Physics, 16, 1479–1489.Find this resource: