When a Solar Geoengineering War Game Exposed How Geopolitics Will Overrule Climate Science
For years, climate researcher Colleen Golja held a quiet, unshakable assumption: when the fate of the entire planet was on the line, rigorous science and evidence-based reasoning would always hold a seat at the decision-making table. That belief shattered one morning last September in London, when she joined the first of a planned series of crisis simulation war games built to head off an unregulated, uncoordinated global dash for a high-risk, experimental planetary cooling technology that could rewrite the rules of life on Earth.
“I thought this would be so simple,” Golja told WIRED. “We could all get on board with ethical, transparent research designed to benefit every human on the planet. Man, was I wrong.”
The simulation, titled Exploring Geopolitical Tipping Points and Climate Cooling Strategies, gathered roughly two dozen cross-sector participants from across the globe: foreign relations experts, active and former military personnel, ex-government officials, and leading climate scientists, all convened in a meeting room at Queen Mary University of London. After casual introductions over coffee and single-serve yogurts, they took their seats around a square table marked with the role they’d been assigned: representatives of national governments, climate nonprofits, Indigenous community groups, and private tech firms. Golja’s official remit was to act as an independent technical advisor, providing scientifically grounded guidance to keep participants from making factually uninformed calls.
The entire exercise was built to stress-test global responses to the growing cohort of experimental technologies grouped under the umbrella term solar radiation modification (SRM). These tools aim to stall or reverse global warming by reflecting a small fraction of incoming sunlight back out to space, most commonly by spraying sulfur dioxide or other engineered aerosols into the stratosphere. The core cooling effect of this solar geoengineering approach is broadly accepted by the scientific community, as it mimics the well-documented temporary global cooling (of 0.5°C or more) that follows massive volcanic eruptions. But its cascading side effects are far less understood, and deeply unequal: models project prolonged drought across sub-Saharan Africa, far more intense Atlantic tropical cyclones, and a catastrophic “termination shock” event if deployment is suddenly interrupted, where global temperatures would spike back to pre-SRM levels with chaotic, nearly unpredictable ecosystem impacts.
Until very recently, SRM was treated as a speculative, niche corner of climate science. That landscape has shifted with extraordinary speed. Private firms have begun pursuing solar geoengineering as a commercial opportunity: last year, US-Israeli startup Stardust Solutions raised $60 million in funding, and in May it published details of a suite of new inventions it claims will give humans the ability to precisely “dial down” global average temperatures at will. In August, Elon Musk – whose company SpaceX already operates heavy-lift launch vehicles and high-altitude flight capacity perfectly suited for SRM deployment – threw his public support behind the technology, posting on X that “space satellites to control temperature and massive geo-engineering will be needed or it will be game over.”
Former UN advisor Cynthia Scharf, who also participated in the simulation, told WIRED that governments will have to make binding choices about SRM governance far sooner than most sitting policymakers realize. “This isn’t a problem 15 years from now. This will all unfold easily within the next decade.”
It was against this fast-moving backdrop that Danielle Young, an assistant professor of international relations at the University of Leeds, designed and ran the London war game. The scenario was framed as an emergency, hastily assembled UN committee tasked with coordinating cross-border responses to fast-moving SRM breakthroughs. The simulation played out across three staggered time markers: the present day, September 2027, and September 2028, with stakes rising at every step as the game’s designers rolled out new, increasingly advanced sunlight-reflecting technology developments. Just like in real life, no player held all available facts: participants were forced to react to ambiguous, partial information that pushed the world closer and closer to full-scale uncoordinated SRM deployment, even when they had no clear picture of what had actually happened, or what rival actors were doing behind closed doors.
The very first session set the tone for everything that followed: distrust and deliberate disinformation dominated every discussion. As participants debated rules for small-scale outdoor SRM testing to try to strike a global binding deal, the players representing the US and China both made it explicitly clear they would never sign any pact that restricted research they framed as being in their “national interest,” nor would they accept a deal that let their superpower rival pull ahead technologically. US Army Major Jeffrey Nielsen, playing the role of a US national security attaché, summed up that stance to WIRED: “Any deal that favors China is far worse than no deal at all.”
Tensions boiled over when the simulation team handed the Russian delegates a classified “intelligence report” claiming a US research plane tied to a secret SRM program had been spotted operating near the Bering Strait. The US delegation – which was, in fact, running unannounced test flights in the scenario – immediately accused Russia of spreading deliberate disinformation. At the same time, the US side grew fully convinced China was running its own secret SRM program, after Chinese representatives attempted to insert a clause into a draft deal that would grant full amnesty for all past unreported SRM experiments. China later privately approached the player representing SpaceX to float the idea of a joint SRM technology-sharing partnership, which the SpaceX roleplayer immediately leaked to the US delegation. The US demanded an immediate international investigation. Finger-pointing escalated until all parties abandoned the talks entirely, with the Canadian representative calling the collapse a “humiliating defeat.”
Nielsen was initially exhilarated after the session, describing his role in blocking a China-friendly global deal as a “huge victory” for US interests within the simulation context. But scenario designer Young dropped a critical reveal after the round ended: China had never actually conducted any secret SRM tests in the scenario at all. Mutual paranoia had led every single delegation to leave the room fully convinced their rivals were sprinting to build out unregulated SRM capabilities as fast as possible.
“What played out that first day is, in my view, exactly what a real-world geoengineering arms race would look like,” Nielsen told WIRED, clarifying he was speaking in a personal, private capacity and not representing the official views of the US government. Most SRM experts agree that only a superpower-level state can secure the airspace, logistics and safety frameworks required to run large-scale SRM aerosol seeding missions, and no major power will voluntarily cede control over the global “thermostat” to a rival. “Any ambiguous action by one actor related to this technology can easily be misinterpreted as hostile intent by the other,” Nielsen noted.
A former British national security diplomat who was not part of the simulation, and requested anonymity to protect professional relationships, drew parallels between this dynamic and early Cold War nuclear arms races, but emphasized a key difference: unlike nuclear weapons, SRM has clear, tangible collective benefits alongside its highly uneven downsides, making it an inherently non-zero sum technology. “The pressure of not falling behind a rival is an enormous driver for R&D,” the diplomat said, “but that is not at all the same thing as rushing to full deployment.” Scharf added that the only plausible path to avoid a catastrophic US-China SRM rivalry would be a fully transparent, science-centered bilateral collaboration between the two superpowers – a outcome she says she cannot fathom happening under a second Trump administration.
Making the problem even worse is a massive, systemic gap in independent SRM research. The technology has long been stigmatized by the mainstream climate science community, which has treated it as a taboo fringe topic. One SRM researcher told WIRED they faced professional ostracization and blocks to academic promotion for working in the space, and eventually left the field entirely.
Golja was one of the very few independent SRM scientific experts in the room, but she was largely sidelined as delegations prioritized geopolitical posturing. She was stunned by how little most participants cared about the actual climate science underpinning the policies they were debating. The rare times delegations approached her for input, it was usually to ask leading questions designed to pull out selective data that justified their national goals, or framed a rival’s proposed action as recklessly dangerous. She built a custom real-time “emulator” tool to demonstrate how different SRM deployment scenarios would alter regional weather patterns and extreme weather risks, and it was almost entirely ignored, even as simulation organizers flooded players with constant alerts about escalating, catastrophic real-world climate disasters that barely registered as a priority for most delegations.
The final shocking twist played out on the simulation’s last day, when participants received alerts that a massive, unreported stratospheric aerosol plume had been detected. Within the scenario, it turned out to be a secret joint US-Saudi test of Stardust’s SRM technology – though most delegations had no way to confirm that at first. Suddenly, independent scientific expertise was a high-demand commodity: delegations wanted guidance on building a global network of satellites and ground-based observation tools to detect secret unauthorized SRM tests as early as possible. A player representing a little-known nonprofit called the Climate Futures Cooperative approached Golja with a large grant offer to fund exactly that monitoring system. “I thought, that’s great, why wouldn’t I jump at this?” Golja recalled.
It soon emerged that the Climate Futures Cooperative was secretly funded by the global fossil fuel industry. Fossil fuel majors and major petrostates like the US and Saudi Arabia have a clear, direct interest in normalizing SRM as a band-aid solution to climate change that lets them keep extracting and burning coal, oil and gas without immediate rapid decarbonization. The NGO spent most of the simulation quietly sowing procedural chaos to derail talks, and only made the grant offer to Golja at the last minute to deliberately discredit her, paint the independent scientific advisor as a compromised actor, and fully fragment the unified voice of evidence in the room. Several delegations – including the US, Russia, SpaceX and Stardust – figured out the NGO’s secret industry ties, but chose to say nothing, as they saw its disruption as aligned with their own strategic interests.
“It was a total mess,” Golja said. By the end of the final session, she was in tears. No binding global SRM governance deal was ever agreed, even after multiple proposals from the EU and Canada focused on shared safety safeguards and open public research. Every draft framework got bogged down in procedural delay tactics and tense zero-sum standoffs over competing national strategic interests. The simulation’s final report, funded by the UK government’s Advanced Research and Invention Agency, concluded that “strategic competition and state sovereignty drove every simulation outcome, not shared climate priorities.”
The simulation’s designers emphasize that their exercise is not a perfect prediction of the future: outcomes are shaped heavily by scenario design and the personal experiences of the players, many of whom did not have frontline diplomatic or national security experience. But its findings have already echoed across real-world global policy even as the simulation was wrapping up. Earlier this year, two conservative US think tanks published separate reports urging the incoming Trump administration to take the global lead on SRM and counter purported Chinese progress on the technology, coining the Trump-aligned slogan of “weather sovereignty.” This month, the Observer in London revealed the UK Ministry of Defense had commissioned a deep tech security consultancy to conduct extensive research into solar reflection technology, and last week the British government published its first official policy paper on SRM, backing limited further research but explicitly ruling out any near-term deployment. Stardust CEO Yanai Yedvab told WIRED that his firm’s global lobbying work is getting far easier by the day, as “the majority of policymakers are already aware of this technology now, which was not the case just a few years ago.” Yedvab added that Stardust will never run large-scale outdoor tests or full deployment before a binding global governance framework is in place, but acknowledged that “when you develop a tool this powerful, as history has shown us, national interests will inevitably shape how it unfolds. We are not naive about that.”
In the months since the simulation ended, Golja has re-evaluated almost everything about her work. She has pinned a handwritten note to the corkboard behind her desk, outlining her new plan for a world shaped by narrow geopolitical self-interest instead of science and reason: she is moving back to the small coastal town where she grew up in Connecticut, to work with local community climate groups to build sea walls and coastal erosion buffers to prepare for the extreme weather that is now unavoidable. “I want to focus on making a difference where I know my work actually has tangible traction,” she said. After what she witnessed in that London meeting room, trying to convince global power brokers to listen to evidence no longer feels like a worthwhile fight. “Now, the work feels a little more personal.”
