The big green button: stratospheric aerosol injection as a geopolitical dilemma during strategic competition between the United States and China, and implications for expanding aerosol injection near-term research

Abstract

This research examines stratospheric aerosol injection (SAI) policy options through a dyadic international relations framework between the United States and People’s Republic of China (China). Deploying large-scale SAI to manage solar radiation presents states with a novel source of geopolitical influence through influencing global climate systems. While multiple political bodies like the United Nations, European Union, United States, China, and India could feasibly deploy SAI without full global consent, the United States and China are powerful enough to deploy large-scale SAI unilaterally. The United States and China currently perceive themselves as locked in “great power competition” with each other which exposes a mutual SAI national security gap and accompanying policy dilemma. Given their divergent global power strategies but mutual global climate interests, this research assesses how the United States and China could compete or cooperate on SAI strategies. This research’s dyadic analysis of four policy scenarios provides three conclusions. First, the United States and China could each benefit from SAI cooperation whether they are cooperating to deter or deploy SAI. Second, SAI cooperation presents a potential political off-ramp from great power competition that aligns with each state’s mutual climate security interests. Third, expanding SAI research and conventional mitigation could support near-term United States and China policymaking regardless of whether they ultimately pursue SAI deployment or deterrence strategies. SAI advocates and critics alike can use these scenarios and conclusions to better discuss SAI as a geopolitical security dilemma.

Introduction: great power competition between the United States and China

This research approaches climate change policy as a function of realist international politics. While climate change is a global phenomenon that warrants global responses, this research assumes that politics between states is a fundamental component of climate policymaking. This assumption includes that large-scale geoengineering—the deliberate deployment of technologies to alter Earth’s climate systems—will similarly function within international politics. This research focuses on the United States and China for two reasons. First, both the United States and China possess the technological, financial, and industrial potential to unilaterally conduct large scale geoengineering [1] (p. 139). Specifically concerning stratospheric aerosol injection (SAI) with a start-up cost of $35 billion, either the United States or China could feasibly initiate unilateral programs as early as 2040 [2]. Second, the United States and China currently see themselves as locked in a long-term geopolitical struggle for international influence [3] (p. vii and IV-2). This struggle, called great power competition, could shape how the United States and China approach climate change and geoengineering.

Great power competition between the United States and China is driven by each state pursuing what its government believes is in its best self-interests. These interests and state self-images are rooted in their national histories and include a vision of 21st century geopolitics [3] (p. vi). Starting with the United States and then discussing China’s perspective best contextualizes climate policy within great power competition.

The United States sees itself as the provider and protector of a relatively stable world order since the Second World War [4] (p. 6 and 38). The United States’ current political perspective begins with the destruction of Nazi Germany and Imperial Japan and then follows through the dissolution of the Soviet Union. The United States took a leading role in what it saw as defending democratic capitalism from violent autocratic ideologies like fascism and Stalinism. In contrast to Imperial great power competition, the United States feels that its liberalist system has prevented devastating wars like those of 1914 and 1937 (1937 was the start of the second Imperial Japanese invasion of China. It is an earlier start of the Second World War than the 1939 Nazi—Soviet invasion of Poland). The United States now sees the world as a rules-based order which uses international institutions, like the United Nations and World Trade Organization, to resolve disputes. This general narrative is common across multiple United States communications like the 2022 National Security Strategy and Joint Concept for Competing.

Looking to the future, the United States seeks to preserve both the current international order and its central position in that order. The United States understands that the growing economic power of other states will require accommodating more of their interests into geopolitics, but the United States sees China as an exceptional authoritarian state attempting to change global norms [4] (p. 8). To the United States, China’s one-party government and economic might present a major threat to international stability. The United States is trying to compete with China by expanding the United States’ global diplomatic, economic, and military partnerships. Yet even in this competitive context the 2022 National Security Strategy specifies that the United States would rather cooperate than compete with China on climate change policy [4] (p. 9).

China has a different historical view than the United States even though the two share many perspectives on great power politics and climate change. China’s modern political narrative spans back to the Xia Dynasty in 2070 BCE [5]. China sees itself as a historical and rightful global power as evidenced by its several millennia of continuous regional dominance. A critical exception to this historical narrative is “the century of humiliation” stretching from the 1839 First Opium War to the 1949 Chinese Communist Party (CCP) establishment of the People’s Republic of China [5]. This period was marked by violent influence over China by various foreign states culminating in the Second World War and Chinese Civil War. China sees its long economic and political growth since the century of humiliation as a rightful process called “national rejuvenation” which China must continue to pursue [5]. This general narrative is common across multiple Chinese communications like the 14^th Five Year Plan, Ministry of Foreign Affairs press releases, and Xinhua official state news agency articles [6, 7].

Looking to the future, China seeks to complete its vision of national rejuvenation by pursuing multiple domestic and international goals. China seeks to forever protect itself from another “century of humiliation” by building alternative economic, diplomatic, and military networks independent from those of the United States [8]. One target date for the completion of national rejuvenation is the CCP’s centennial anniversary in 2049 [6] (p. 3). While Chinese policy statements only address competition with “the strong enemy” without naming the United States directly, China sees the United States as a major threat to national rejuvenation due the latter’s military and diplomatic partnerships with Taiwan, South Korea, and Japan [9] (p. V and 47), [10]. China is trying to compete with the United States by developing its domestic economy and presenting a credible geopolitical alternative to the current system. Yet even in this competitive context the National Climate Change Adaptation Strategy 2035 specifies that China would rather cooperate than compete with the United States on climate change policy [11] (p. 47).

Comparing these two national narratives highlights the importance of assessing geoengineering through a dyadic framework and security lens. The United States calls China its “only competitor with both the intent to reshape the international order, and, increasingly, the economic, diplomatic, military, and technological power to do it [5] (p. 24).” China calls the United States “the strong enemy” which has “contained, encircled and suppressed China in an all-round way, bringing unprecedentedly severe challenges to China's development [8] (p. V and 47), [12].” Yet at the same time as these adversarial statements, the United States calls climate change a “potentially existential [risk] for all nations” while China calls climate change a “serious threat […] to the present and future survival and development of mankind [4] (p. 9), [11] (p. 4).” This dichotomy of mismatched power interests and shared climate concerns raises this research’s questions. How could the United States and China approach geoengineering during great power competition? How could either state anticipate, respond to, or pre-empt the other’s geoengineering policies? Are there possible win-win geoengineering policies that contribute both to the United States’s rules-based order narrative and China’s national rejuvenation narrative?

Methodology

This research focuses on SAI because it is the leading geoengineering technique likely to provide cost-effective, near-term results to cool down Earth’s atmosphere [13] (p. 3). SAI is a solar radiation modification technique of spraying chemicals into the atmosphere to artificially block sunlight and delay climate impacts. Balancing artificial reflection against greenhouse gas heat retention would slow global warming, thus buying more time for other mitigation and adaptation strategies [13] (p. 4), [14] (p. 4–6). There are multiple engineering conceptions of SAI which vary in their chemical and delivery methods. Some programs envision conventional aircraft spraying Sulfate, while other programs envision autonomous balloons spraying different chemicals [15]. At an estimated cost of $18 billion per year per 1°C in effective cooling, SAI presents an attractive policy option relative to potential trillions of dollars in yearly global climate change damages [15, 16].

Previous research examines a risk-versus-risk framework that a state (State A) could use to decide whether to deploy SAI, usually emphasizing direct environmental benefits and risks [17] (p. 2476), [18]. This research instead examines a threat-based framework that State A’s geopolitical adversaries could use to decide how to respond to State A’s potential deployment. Regardless of exact SAI technologies, a critical facet of great power SAI policymaking will be whether it presents any threats or opportunities to United States and Chinese vital national interests. A state’s vital national interests include protecting its homeland and citizens, protecting its economic prosperity, and protecting its national way of life [3] (p. vii). This research focuses on SAI impacts on these three vital national interests to remove minor considerations which are unlikely to steer policymaking.

Additional assumptions are necessary to scope this research. Following a threat-based approach, this research assumes that either the United States or China could unilaterally deploy SAI. Another assumption is that the United States or China would deploy large-scale over small-scale SAI due to the technique’s up-front costs and diminishing effectiveness below 50% of the global effects mass [15]. Rather than examining many small SAI programs targeted at localized cooling, this research examines “the big green button” as an acute security concern for powerful states (Fig. 1). This research also assumes that the SAI deployer does so as a sincere effort to cost-effectively protect its own vital national interests. This research still examines how other states could interpret the deployer’s motivations as sincere or malicious, but assuming earnest deployment provides clearer deterrence calculations.

For the remainder of this research’s bilateral analysis, “State A” refers to a great power state deploying large-scale SAI for global cooling effects. “State B” refers to the other state. Either the United States or China could be State A or State B. To keep this research within a dyadic framework rather than a global policy analysis, “unilateral” deployment is not strictly so and includes latitude for State A to have the consent of some third-party states. This exception does not affect the dyadic analysis while those additional partners are not powerful enough to fundamentally alter State B’s options. This methodology provides an appropriate framework to assess United States and Chinese SAI policy options without expanding the research to global analysis.

SAI threats and opportunities

Large-scale unilateral SAI by State A imposes four distinct impacts on State B. Three of these impacts are largely security threats to State B: termination shock, environmental damage, and SAI weaponization. The fourth impact category, impacts to State B’s geopolitical influence, includes both threats and opportunities to State B’s vital national interests.

The first SAI impact on State B’s vital national interests by State A’s unilateral deployment is the threat of termination shock. SAI is a fundamentally temporary solution because it masks global warming without reducing or reversing greenhouse gas emissions [19] (p. 26). To offset current warming, aerosols must be re-injected on much shorter timescales than the natural sequestration of greenhouse gases out of the atmosphere (6 months to 3 years versus decades and centuries) [20] (p. 6), [13] (p. 14). Rather than outright replacing conventional decarbonization costs, SAI only provides more time to spread those costs out. An essential threat of any adversary SAI deployment is this effect, called mitigation deterrence, crowding out conventional climate policies in the unfounded belief that SAI “solves” climate change [21].

If states, corporations, or any other bodies use successful global SAI to justify continuing their individual greenhouse gas emissions, then any cessation of SAI could suddenly jump the world to a much higher energy imbalance and resultant accelerated warming [14] (p. 6). This effect, called termination shock, could accelerate climate impacts into a shorter and more dangerous timeline on unprepared global systems [14] (p. 6). While the likelihood of mitigation deterrence is currently debatable, State B has a vested interest in determining whether State A might fundamentally alter the nature and pace of climate change [22]. This is especially concerning to State B if termination shock increases the demand on its security and weather disaster aid resources.

The second SAI impact on State B is potential direct environmental damage. The nature and scale of these impacts is unclear, but State B will want to understand how State A injecting billions of kilograms of anthropogenic chemicals into the stratosphere will impact State B or its allies [13]. SAI could severely damage global atmospheric, hydrological, and ecological systems and then damage the societies and economies that depend on those systems [19] (p. 39). Though unlikely in many SAI deployment scenarios, accidental cooling overshoot presents similar risks that could impact State B [22].

The third SAI impact on State B is the threat of State A weaponizing SAI whether directly or indirectly. Direct weaponization is less plausible because a direct SAI attack will still be attributable to State A as dozens or hundreds of vehicles conduct the attack. SAI offers State A no distinguishing direct attack capabilities over existing strategic weapons and thus falls outside the scope of this research. The more plausible threat is that State A can indirectly weaponize SAI by concealing malicious intent within its otherwise sincere injection strategy. With sufficient information and engineering, State A could deploy SAI in a manner that appears benign but secretly preferences certain climate outcomes over others [23] (p. 115–117). Example concerns include State A prioritizing Himalayan or Arctic re-glaciation over Southeast Asian monsoon stability [24].

The fourth SAI impact is the potential for State A to leverage large-scale SAI to displace State B’s geopolitical influence. If State A’s SAI program succeeds in reducing global climate impacts and cost-effectively buying time for other policies, then State A could claim that it “saved” the world from worst-case climate scenarios [1] (p. 148). State A could align this narrative with its other great power competition methods. For example, State A could acquire global diplomatic and information influence among populations relieved to see decisive climate action in the face of failing conventional approaches [1] (p. 139), [25] (p. XII), [26] (p. 41–44). State A could acquire even more influence if it deploys SAI while publicly rebuffing State B’s SAI deterrence. Like narrative struggles during the Cold War, State A could claim that it is once again the guarantor of global stability or claim that it has provided a better alternative to a stalled international system.

State A could also leverage SAI power to coerce other states or justify applying extraneous pressures. If State A asserts that it is protecting others from worst-case climate change then it could demand cost-sharing or political concessions in return. In extreme circumstances, State A could brandish the risk of termination shock to assert or enforce other policies. State A could legitimize these malicious interventions under the guise of enforcing necessary decarbonization to stave off termination shock [27] (p. 40). The ongoing trade conflict over electric vehicles provides a glimpse at how these conflicts might unfold during SAI great power competition [28, 29]. State B will seek to counter-act or pre-empt State A’s prestige victory rather than accept State A’s SAI preeminence.

There is one critical opportunity for State B’s geopolitical influence within State A’s unilateral deployment. Considering climate change as a shared United States and Chinese existential threat with the potential to cause trillions of dollars in damage, State B could “free ride” on State A’s SAI to receive full climate benefits without paying any SAI costs [30]. Deploying SAI will incur various geopolitical influence costs on State A to provide global cooling. State A will have to create sufficient domestic political consensus for deployment. State A will have to navigate international disapproval of unilateral deployment, even as multilateral efforts fail [14] (p. 19). State A will have to spend tens of billions of dollars up-front and annually to sustain SAI for an indeterminate period. In the meantime, State B could spend its resources pursuing other great power competition goals while reaping the benefits of reduced climate change impacts.

It is important to note that the United States and China could prioritize defending their national sovereignty in relation to SAI impacts. In addition to quantifiable technical threats that State A’s injection strategy imposes on State B’s interests, State A imposes an inherent sovereignty threat by devaluing State B from the SAI decision process. The exact nature and scale of State A’s deployment impacts on State B could be less important than whether State B is willing to cede such important decisions to one of its primary geopolitical competitors. Table 1 displays all four impact categories across State B’s vital national interests.

Four SAI policy scenarios

Combining the four impacts of State A’s unilateral deployment on State B’s vital national interests provides a starting point to evaluate United States and Chinese SAI policy options. These options can be categorized into two binary sets for a total of four policy scenarios: competition versus cooperation and deterrence versus deployment. Competition includes the United States and China pursuing separate SAI policies during great power competition. Cooperation is an SAI policy agreement between both states. Deployment is the initiation of aerosol injection driven by states’ risk-versus-risk SAI calculus. Deterrence is one state using threats or concessions to compel another state to abstain from deploying SAI. These scenarios are not static, and future great power SAI could shift between individual scenarios over time as either state adjusts its SAI policies. Each scenario presents multiple risks (red) and opportunities (green) to the United States and China regardless of which is State A or State B as displayed in Table 2. This section describes each scenario in depth based on the SAI impacts from Table 1.

Scenario 1: compete while deterring SAI, “Great Power Status Quo”

In Scenario 1, the United States and China attempt to deter each other from deploying SAI as a general facet of great power competition. In this research’s terminology, both the United States and China act as State B while protecting themselves from the possibility that the other becomes State A. A historical analogue for this scenario is 1950s and 1960s United States and Soviet Union mutual nuclear deterrence.

This scenario presents multiple risks. First, deterring SAI without expanding conventional mitigation sustains the innate risk of current worst-case climate change projections. Deterring SAI while continuing great power competition and status quo climate policies just delays the geoengineering question without providing a credible alternative. A subsequent long-term risk is multipolar SAI competition as growing climate impacts incentivize more states to consider deployment, thus expanding general deterrence requirements [17] (p. 2478). Even under the most optimistic emissions reduction assumptions, the potential for catastrophic climate impacts suggests that SAI deterrence could be a decades or even centuries-long policy [25] (p. 30–32).

SAI deterrence presents a substantial policy dilemma. Deterrence is a function of threats and concessions to compel an actor away from an action, and both methods are poorly suited to SAI deterrence. Assuming that climate impacts are a core reason for states to deploy SAI, concessions would include very expensive options like State B paying for State A’s climate adaptation or damages. These options are unlikely to be feasible given their several-trillion-dollar price tags [31, 32]. Any in-kind concessions or repayments from State B outside of the climate domain would need to be similarly valuable to State A. Either type of concession is less likely to be politically feasible during great power competition as neither state will want to strengthen its adversary, especially at high cost to itself.

An additional consideration for State B is whether a delay between decarbonization costs and its climate mitigation payoff creates a mitigation abandonment window. If decarbonization costs are visibly greater than decarbonization benefits for a long enough period, State A could still trend towards SAI deployment despite decarbonization sunk costs. Alternatively, State B could abandon decarbonization aid outright for the same reason of expensive mitigation programs outweighing intangible dividends. Either case presents a costly risk of expensive decarbonization policies failing to support SAI deterrence (Fig. 2). State B will seek political assurances from State A to hedge this risk before financing any decarbonization concessions, but State A cannot know how future climate damages to its vital national interests might change its future SAI policy decisions.

SAI deterrence threats are equally risky. Threats rely on State B’s capability and credibility to follow-through on retaliation if State A acts in the undesired way (unilaterally deploying SAI). State B must also communicate its retaliatory threats early and effectively enough for State A to include those threats in its decision-making. However, if State A sincerely believes that SAI is a necessary tool to protect itself from severe climate damages then deterring it could have an unobtainably high deterrence threat threshold. The unclear international legality of large-scale SAI could also embolden State A rather than help deter it [33] (p 119–124). This high deterrence threshold could also be reinforced by increasingly dire climate warnings even if those warnings are only intended to advocate conventional mitigation [34]. In the process of communicating with each other over deterrence threats, State B could eventually be cornered into communicating uncredible and therefore ineffective threats. State B will be challenged to find an exact combination of threats that makes SAI more risky than other climate change options yet are reasonably credible within State B’s shared climate worries.

For example, envision State A proposing multilateral SAI at successive United Nations conferences from 2028 through 2035. As the proposals fail to garner global consensus, State A increasingly points to floods and droughts in its homeland to justify unilateral deployment. State B attempts to deter State A by threatening to sever economic ties with State A and exclude it from multiple military, diplomatic, and trade organizations. State A accedes to the threat for a few years, but eventually rebuts after a historic extreme weather event damages its largest cities (a “climate Chernobyl”) (the 1986 Chernobyl nuclear reactor disaster was large enough to impact Cold War geopolitics beyond nuclear energy policy). As State A formalizes its intent to unilaterally deploy SAI no later than 2045, State B finds itself in a dilemma. If State B backs down, then it cedes SAI decisions to State A and accepts all the impact risks that come with that. If State B instead escalates its threats, then it risks being rebuffed as hypocritical or otherwise following-through on severe threats like violent retaliation.

If State B threatens to shoot down State A’s SAI aircraft, will State A even believe the threat given State B’s decades of equally dire climate change warnings? How will third parties and State B’s own domestic political groups, especially SAI advocates, react to this escalation and lost opportunity to free ride on State A’s resources [18, 26] (p. 33), [35] (p. 131–133), [36]? Has State B prepared sufficient legal justification and domestic consensus to initiate armed conflict over SAI? The crux of SAI deterrence is the potentially fatal mismatch between trying to deter State A from doing something it sincerely believes is in its own protective best interests. This mismatch is even greater if State A has credible evidence to argue that it is acting in State B’s best interests, and the breadth of climate warnings by each government could provide that evidence.

Competing while deterring SAI does present one benefit to each the United States and China. This scenario most closely aligns with the current great power competition status quo because neither state is publicly initiating SAI programs nor is likely to support a sudden large-scale attempt. This scenario allows both states to forego any of the complicated political costs of deploying SAI or cooperating to deter it. Both states might simply perceive competition as their best option while the geoengineering security situation continues to develop.

Scenario 2: cooperate to deter SAI, “Decarbonization G2”

In Scenario 2, the United States and China formally commit to a mutual deterrence posture and cooperatively deter SAI either from each other or a third party. Each state consciously functions as State B and understands the other’s sincere intent to stay State B. Rather than a G7 for economic coordination, the G2 forms for environmental coordination. A historical analogue of this scenario is the 1968 Nuclear Non-Proliferation Treaty including the United States and Soviet Union.

This policy scenario maintains the same inherent climate change risks as Scenario 1. It also maintains the subsequent potential for future multipolar SAI competition in the face of increasing climate impacts. This scenario has the same deterrence risks as Scenario 1, but with the added benefit that the United States and China cooperatively deterring third parties could more credibly succeed than either state acting alone. This is true for both deterrence threats and concessions.

As a security baseline, deterrence threats against third parties would be more credible than between State A and State B. Unlike the United States and China trying to deter each other, there are few states or plausible coalitions that could resist combined pressure from two of the world’s most powerful states.

The United States and China could also combine their economic resources to provide global concessions like expanding conventional mitigation. Though these costs would still range in the trillions of dollars, the United States and China could plausibly share the costs in ways that align with both of their current economic and geopolitical efforts. For example, the United States could expand its existing goal of mobilizing $100 billion yearly to global mitigation, decarbonization, and adaptation policies to developing countries [37] (p. 62). China could simultaneously expand its own industrial and technological support to developing countries to synergize global mitigation (perhaps with matching United States investment). Expanding mitigation far beyond current policies could more credibly alter State A’s SAI decision criteria as its fear of climate risks shrinks.

Expanding conventional decarbonization is already a cornerstone of international climate change advocacy [38]. This cooperation is feasible exactly because both governments have specified their preference to cooperate on climate change mitigation even during great power competition [4] (p. 9), [39] (p. 374), [40], (p. 9). However, framing large-scale SAI as a national security threat provides a different reenforcing logic for mitigation and adaptation. Earnest SAI debate exposes the profound implications of a future geoengineering paradigm in ways that could attract action even from bodies that are unconcerned about direct climate change impacts. Furthermore, exposing the political challenges of SAI deterrence can help persuade powerful states into providing more global support to avoid worst-case climate scenarios. In the most optimistic case, “Decarbonization G2” cooperation could even galvanize global climate efforts as decisive climate action seems more achievable than before.

The United States and China tangibly expanding conventional climate change policies to cooperatively deter SAI could also support their geopolitical narratives. The United States could claim that its international order has once again successfully resolved an international conflict. China could claim that its economic might and diplomatic influence have realized national rejuvenation, especially if the world sees climate progress before 2049. Should positive SAI deterrence succeed in guaranteeing safer global climate change pathways, this scenario could even provide a cooperative off-ramp from great power competition as each state enjoys a mutual prestige dividend. A credible path towards “great power cooperation” could be attractive to both the United States and China.

Scenario 3: compete to deploy SAI, “Race to the Stratosphere”

In Scenario 3, the United States and China adopt competitive SAI policies with at least one pursuing unilateral SAI. One acts as State A and State B must decide between pre-emptive SAI, concurrent SAI, or SAI deterrent approaches. A historical analogue for this is the 1960s United States and Soviet Union arms race for a dominant nuclear arsenal.

SAI competition between State A and State B invokes a new dilemma as State B identifies a “unilateral SAI security gap” driven by State A’s potential to win an SAI arms race. A security gap is a national security threat that a state is insufficiently prepared to defend itself from. One politically damaging outcome for State B is State A successfully deploying large scale SAI first and thus dominating any geopolitical influence that SAI provides. However, State B will face all the deterrence challenges posed in Scenario 1. State A deploying SAI to sincerely protect its own vital national interests against climate change, while conscious of State B’s substantial free rider benefits, could easily see past State B’s deterrence efforts as not credible. This challenge exposes an SAI security gap that State B might seek to close as soon as it identifies the possibility of State A’s unilateral deployment, even if the actual deployment is years or decades away.

It is important to describe the SAI security gap in depth to understand how it could impact both states’ behavior. Because the United States and China are each powerful enough to overcome SAI’s technical and industrial hurdles, the domestic political consensus for deployment could be decisive in an SAI race. The key advantage might be in deciding to deploy rather than building the capacity to deploy. Multiple factors could provide State A deployment decision advantage over State B. For example, the United States could have decision advantage due to its international climate security concerns or fossil fuel monetary interests [22, 41]. The United State is currently the world’s largest oil producer, and SAI presents an opportunity to protect that industry by delaying worst-case climate impacts [42]. Alternatively, China could have decision advantage due to greater climate vulnerability and a more permissive legal-political environment [17] (p. 2478), [18, 39] (p. 18). China is the world’s second largest oil consumer, and SAI presents an opportunity to align climate policy with 2049 national rejuvenation without compromising any economic development goals.

In either case, State A’s decision advantage creates a dangerous window for State B when SAI deterrence seems like a losing option, yet deployment is politically unacceptable. Therefore, great power competition might color SAI competition much sooner than potential deployment windows as both the United States and China seek to protect themselves from being caught as a vulnerable State B (Fig. 3). The United States is already debating funds for SAI detection platforms “to prevent strategic surprise,” which suggests an initial awareness of this security gap [43, 44]. This research did not identify similar Chinese programs, but there is potential for protective actions by one state to be perceived as escalatory by the other: “is State A researching SAI to protect itself, or to understand how to prevent State B from protecting itself?”

Counterintuitively, closing the unilateral SAI security gap could increase geopolitical risks as each state reacts to the other’s preparations. This in turn could accelerate SAI to impact geopolitics much sooner than SAI would physically be deployed, increasing each state’s relative urgency of initial SAI policymaking and any research that informs policymaking. Time pressure alone could increase all SAI risks as policy debates occur faster than SAI researchers can generate more accurate technical and experimental information.

The asymmetry of decision advantage factors might caution both states from initiating an SAI arms race. While the other state’s SAI position is ambiguous, domestic political processes can continue to function normally within the great power competition status quo. However, overt pro-SAI action by State A could instead give State B the political consensus it needs to act quickly against a new strategic threat. Factoring in the SAI security gap and long-term geopolitical influence risks of SAI deterrence, State B could attempt to develop its own SAI capability to pre-empt State A or otherwise deploy SAI at the same time as State A.

This decision advantage SAI arms race exacerbates termination shock and environmental damage risks. Two states rushing to pre-empt the other’s deployment could engineer their SAI strategies less prudently while trying to prevent its adversary’s SAI preeminence. The decision-making risk-versus-risk process could be a tempting place to shave off deployment time in pursuit of pre-emption, especially if policymaking urgency outpaces prudent SAI research. Both states might make riskier decisions to close a great power competition security gap than they would otherwise make approaching SAI as a decontextualized climate policy.

A unique risk of concurrent, competing large-scale deployments by State A and State B is a “competing storms” situation [45]. The world could find itself in an unmanageable stratospheric three-body problem as uncoordinated great powers struggle to control SAI. Single-actor large-scale SAI is already a risky proposition of unprecedented direct human control over climatic systems. Adding other large-scale actor variables can only complicate the sensitive SAI equation and exacerbate technical risks.

Aside from the potential climate change cost savings of successful SAI in Scenario 3, there is potential opportunity for State A to expand its geopolitical influence by achieving a global prestige victory. Successful SAI could be a powerful contributor to climate action as an arena of great power competition [46]. Even within a sincere deployment to protect itself from climate damage, State A could be attracted to SAI’s power potential. State A could directly align SAI with its geopolitical narratives or expand SAI coercion into a new form of diplomatic power. State A must still carefully consider State B’s reaction, especially given the unclear winner of an SAI arms race decided more by political agility than technical capability.

Considering all these risks and opportunities, Scenario 3 might be the worst-case SAI policy approach for both the United States and China because it adds considerable new risks while compounding the existing risks of other scenarios. State B stands to suffer multiple damages to its vital national interests while deciding whether to risk deterrence failure, deployment failure, or three-body SAI failure. Any State B free rider benefits will also be tainted by State A’s prestige victory, and doubly so if State A succeeds while publicly rebuffing State B’s deterrence. However, State A also incurs several risks in this scenario that it would not otherwise experience during SAI cooperation or mutual deterrence. State A must develop its SAI program under additional geopolitical pressure, possibly at great cost to its international legitimacy and domestic political stability. State A must also manage innate SAI risks for the full duration of a deployment while watching others free ride its program. State A must create multiple precise circumstances to reliably “win” an SAI arms race.

An unfortunate aspect of Scenario 3 is that it only takes one great power to initiate the SAI arms race, so it might be both the most dangerous and likely of this research’s four scenarios. Even if early great power SAI policymaking is defined by deterrence, states can change their policies as political or climate conditions change. State A building any amount of SAI capacity could initiate an irreversible chain of suspicion and self-defense that starts an SAI arms race.

Scenario 4: cooperate to deploy SAI, “Aerosol G2”

In Scenario 4, the United States and China cooperatively deploy SAI as a global climate change mitigation technique. Both act as State A within cost and power sharing arrangements. A historical analogue for this scenario is the 1972 United States and Soviet Union space exploration cooperation agreement.

Cooperative SAI presents a unique opportunity for each state to receive the economic benefits of SAI while mitigating technical and governance risks [13] (p. 39). A bilateral deployment would involve both the United States and China cooperating to deploy SAI even without full global consensus. In contrast to unilateral deployment, United States and Chinese SAI cooperation could mutually optimize injection strategies to minimize environmental risks, select amicable goals to remove weaponization risks, build political consensus to remove decision advantage risks, provide redundant funding to mitigate termination shock risk, and globally legitimize the program.

More importantly, this option could reasonably align with United States and Chinese respective policy statements to increase climate cooperation during overall great power competition [5] (p. 24), [11] (p. 47). In the most optimistic conception of Scenario 4, cooperative SAI could provide an off-ramp from great power competition as each state satisfies its respective geopolitical narrative without infringing on the others’ vital interests. The United States could feel successful in integrating China into a more secure cooperative international system, especially if bilateral deployment expands into mini-lateral deployment with other partners. China could feel successful in achieving national rejuvenation through diplomatic and economic means. Though Scenario 4 may appear implausible within current competition, the unprecedented scale of climate change risks could equally incentivize unprecedented great power cooperation. The United States and China were military partners as recently as 1984 which provides a longer perspective on political possibilities between 2025 and 2065 [40] (p. 9).

The emergent risk of these various SAI opportunities is the difficulty of creating acceptable SAI governance tools between two competing great powers [13] (p. 40), [47]. As with Scenario 3, there could be a long period of time during which State A wants to deploy SAI and could even prefer to do so cooperatively, yet State B still considers deployment unacceptable or premature. State A might be willing to delay its deployment overtures for a few years while trying to persuade State B into cooperation, but the unclear duration of this period complicates that option. It is unreasonable to assume that State A will delay its SAI indefinitely if worsening climate change damages its vital national interests, so a long enough mismatch window will trend State A towards unilateral deployment and thus Scenario 3.

One option for State B is to accept State A’s deployment as a form of implicit SAI cooperation. However, that option could also trend more towards Scenario 3 as State A seeks cost-sharing and State B seeks to mitigate State A’s SAI geopolitical influence impacts. Sustaining Scenario 4 requires both states to identify SAI deployment as an acceptable option roughly at the same time, plausibly from a shared “climate Chernobyl” like a global food price shock or surge in climate migrants. Otherwise, State B must choose to cooperate from a long-term geopolitical perspective despite its environmental SAI wariness.

It is essential to note that some innate SAI risks cannot be engineered out of an SAI deployment: aerosols do not restore the Earth’s atmosphere to its thinner, pre-industrial composition. Even during cooperation, both states could seek reassurances against mitigation deterrence. Rather than influencing global actors towards conventional decarbonization, emerging “Aerosol G2” cooperation could trigger mitigation deterrence long before the first SAI payload is even delivered. This version of mitigation deterrence would function closer to “anticipation shock” than “termination shock,” which creates an additional risk layer for the United States and China to juggle. In the worst-case of SAI anticipatory mitigation deterrence, even combining United States and Chinese power could still commit humanity to a dangerous long-term SAI paradigm with no clear off-ramp. SAI as a source of geopolitical influence might legitimize the “Aerosol G2” to more aggressively enforce global decarbonization policies, but this possibility generates poignant international security questions whose analysis is beyond the scope of this research.

Discussion: policy options that support each scenario

A great power competition dyadic framework between the United States and China reveals complex, competing risks and opportunities in all four policy scenarios. The asymmetry of geopolitical influence dynamics, climate change impacts, and SAI geopolitics exposes a United States—China mutual security gap and accompanying policy dilemma. Either state could function as State A or State B, yet there is no clear optimum SAI strategy for either. Within great power competition brushing up against the climate crisis, SAI presents challenges that could demand nuanced geopolitical solutions rather than technical solutions.

Describing these four scenarios in depth highlights two important policies that the United States and China can already pursue regardless of their future SAI decisions: (i) expanding SAI research, and (ii) expanding conventional climate change mitigation. Both policies could support future United States and Chinese SAI scenarios regardless of if they are competitive, cooperative, deterrent, or deployment policies.

Whether to increase SAI research is a contentious issue, especially with respect to outdoor experimentation [48]. However, expanding research supports future SAI decisionmakers without biasing them towards or against deterrence or deployment [19] (p. 111). As highlighted in Scenario 3’s discussion of the SAI security gap, SAI researchers can mitigate potential policymaking time pressure by accelerating and expanding research even if plausible large-scale deployment is further in the future.

Scientifically, expanding SAI modeling and outdoor experimentation would inform states on injection detection, chemical dynamics, direct environmental impacts, and secondary impacts [13] (p. 16). Politically, SAI research can refine the legal and governance frameworks to either detect and deter SAI or support an equitable future deployment. Combining these two sets of information would allow State A and State B to both make more informed SAI policy decisions through a more detailed understanding of program opportunities and risks. Expanding near-term SAI research, to include outdoor experimentation, helps close the SAI security gap whether the United States or China ultimately chooses deterrence or deployment. Cooperative SAI research could also mitigate the geopolitical risks of either state interpreting the other’s research as a security threat.

This research’s framework provides a geopolitical starting point for further SAI security research. First, future research could apply this dyadic framework to other powerful international actors like the European Union, Russia, and India. The European Union and Russia are important stakeholders in any potential Arctic SAI, and India is an important stakeholder in any potential Himalayan SAI. All three could reasonably influence global SAI decisions. Second, future research can expand this four-scenario framework from two states to two coalitions of states for a better global analysis. In lieu of State A and State B, such research could analyze policy scenarios for Coalition A (SAI deployment), Coalition B (SAI deterrent), and Coalition C (policy-neutral) [49]. Such research will be difficult as the exact members of any one coalition could drastically alter scenario outcomes, but the potential urgency of initial SAI policymaking could justify the resources for professional SAI game research. Third, future research can explore mechanisms to influence states towards any of this research’s four SAI scenarios. One research technique would be to add large-scale geoengineering to strategic-level climate and diplomatic wargames [50].

Expanding conventional climate change mitigation is another SAI-neutral policy because it both decreases the incentive for deployment and decreases the likely duration and injection mass of any future deployment. The fundamental motivation for State A’s SAI in this framework is to mitigate climate impacts via global cooling to buy time for more conventional policies. Though more costly, other mitigation policies accomplish this same purpose without the technical or governance risks of SAI [13] (p. 11). That China invested $676 billion, and the United States invested $303 billion, in renewable energy infrastructure in 2023 is evidence that both states sincerely want to solve the climate crisis [51]. Expanding conventional mitigation aligns with these policies while reducing SAI risks in all four potential scenarios. While this policy is unlikely to deter SAI outright, improved emission pathways could positively reduce the scale and duration of any deployment.

As identified in Scenario 2, a risk of cooperative decarbonization is its ultimate failure if states try to deploy SAI even in lower climate change impact situations. State A and State B could end up in the same SAI dilemma just a few trillion dollars later, and both states will want to hedge that risk. Still, the floundering progress of international climate cooperation implies a need to galvanize global effort [25] (p. XXII). An earlier discourse to close the SAI security gap could serve as a catalyst for the United States and China to lead necessary global change. This research provides an alternative, reinforcing logic to conventional mitigation: climate change could drive powerful actors towards unilateral SAI, which could be very difficult to deter. States that wish to deter SAI might be better off leveraging cooperative concessions earlier than leveraging competitive threats later.

Conclusion

As the body of geoengineering technical literature grows there are many future geoengineering paradigms worth examining through a threat-based security framework. The potential for powerful actors like the United States and China to deploy large-scale SAI reveals a complex web of how great power competition could shape geoengineering policy. SAI’s unique position in the climate crisis creates a security gap and accompanying policy dilemma for each state. State A will have to balance its dire climate change perceptions with the geopolitical risks of SAI deployment. State B will have to balance equally dire climate change perceptions with the geopolitical risks of SAI deterrence. Both states stand to lose from SAI competition whether deploying or deterring, yet cooperation could prove elusive while both states hold different SAI perceptions. Both states stand to benefit from great power cooperation regardless of whether they pursue deterrence or deployment together, yet cooperation incurs its own set of governance challenges. There is no overt optimum SAI strategy for either state to pursue during great power competition.

Fortunately, both the United States and China can support their future SAI decisions through two SAI-neutral policies. Expanding near-term SAI research (both scientific and security-focused) and expanding conventional climate change mitigation could support their vital national interests regardless of their future SAI preferences. Whether they eventually adopt SAI cooperative-competitive or deterrence-deployment approaches, better understanding of SAI impacts and lower anticipated climate damages will better support future decision makers. Instead of a new decarbonization obstacle, the expanding SAI discourse presents an opportunity to reinvigorate conventional climate action.

Finally, this research suggests that future SAI research can embrace the geopolitical implications of such a largely unprecedented technology. Advocating for or against SAI as a discreet climate issue separated from geopolitics misses critical SAI risks and opportunities. If the goal of SAI discourse is to inform policymaking, then framing SAI as a security issue is a key method to discussing its broad implications. SAI literature can reflect these realist international relations perspectives to help states prevent a geoengineering future or else help states navigate it.

Acknowledgements

The author thanks Dr Nowowiejski and the 2024 Art of War Scholars for their support during this research. The author also thanks his spouse, Paula Skandis.

Author contributions

Jeffrey Nielsen (Conceptualization [lead], Formal analysis [lead], Investigation [lead], Methodology [lead])

Conflict of interest: The author has no relevant financial or non-financial conflicts of interest to declare. The views expressed here are those of the author and do not necessarily reflect the views of the United States Army or Department of Defense.

Funding

The author did not receive any funding for this research.

Data availability

Available from the author upon request at [email protected] or [email protected].

Ethics statement

 

References