Controversial schemes to control sunlight and cool the planet get a serious look as the toll from climate change mounts.
By Mark Matthews
Aerospace engineer James T. Early built a 30-year career at the Lawrence Livermore National Laboratory working out big ideas. Among them: the use of pulsed lasers to knock space debris out of orbit, and a giant telescope—powerful enough to detect planets in distant solar systems—with lenses that roll up to fit inside a rocket and then spread out when launched in space. Trained at MIT, Caltech, and Stanford, Early drew inspiration from the science fiction of Isaac Asimov and Arthur C. Clarke. When his wife wanted to try her hand at the genre, he outlined a debut plot: terraforming Venus. An immense sun-blocking shield, he theorized, could cool the planet over time and turn its scorching, desiccated surface into a human habitat.
Early wondered if a scaled-down sunshield could solve the real-life problem of a warming Earth caused by carbon-spewing power plants and vehicle exhaust. In three terse pages published in the Journal of the British Interplanetary Society in 1989, he spelled out a “conceptually simple method” for offsetting the heat-trapping greenhouse effect and cutting global temperatures by 2 degrees Celsius (3.6°F). A glass shield could be fabricated on the moon out of lunar soil and dispatched, using an electromagnetic accelerator, to the first Lagrange point (L1), a million miles from Earth, where the gravitational pull of the Earth and sun cancel each other out. He hazarded a price tag of $1 trillion to $10 trillion—“enormous,” yes, but perhaps much lower than the economic impact of the greenhouse effect.
Roger Angel, an esteemed astronomer at the University of Arizona, built on Early’s idea in 2006. He conceived a giant space cloud at L1 comprising trillions of thin reflective satellites, each a meter wide and weighing 1 gram. These little “flyers” would be kept aligned with the sun by solar sails, which are powered by photons of sunlight bouncing off a mirrored surface. Manufactured on Earth, the satellites would be sent aloft to L1 in stacks of 800,000 using electromagnetic acceleration and ion propulsion. They would form a cloud some 60,000 miles long and 2,000 miles across, weighing 20 million tons. Angel’s detailed six-page description in the Proceedings of the National Academy of Sciences estimated a cost of “a few trillion dollars.” The idea, he wrote, might be desirable if “dangerous changes in global climate were inevitable.”
Early and Angel were each ahead of their time in combining futuristic space technology with the science of preventing catastrophic climate change. But their vision of a vast sunshield looks a bit more realistic now, amid such advances as a fledgling space tourism industry, reusable rockets curbing the cost of space launches, and growing interest in space-based manufacturing and moon and asteroid mining. Meanwhile, as the effects of climate change become alarmingly clear and warnings by the Intergovernmental Panel on Climate Change grow more precise and ominous, a growing number of scientists, engineers, graduate students, and members of Congress are looking to geoengineering as a possible when-all-else-fails technical fix for an overheating planet.
Enter Space Bubbles, a 2022 variation on the sunshield idea from a team of MIT engineers. Led by Carlo Ratti, director of the university’s Senseable City Lab, the group envisions giant space platforms—“rafts”—composed of stuck-together inflatable spheres of thin-film silicon or another molten material, positioned slightly closer to the sun than L1. The spheres would be manufactured using “space-based fabrication methods” not clearly defined. Like the Early and Angel concepts, the rafts would block 1.8 percent of the sun’s rays—an amount sufficient to prevent dangerous warming. The cooling could begin by the end of this century, “when the most severe consequences of climate change are currently predicted,” the team says. As to price, “Our initial estimations suggest that the Space Bubble Raft will have lower mass-density than Roger Angel’s proposal and might thus be more cost-efficient,” Ratti tells Prism.
Space Bubbles brings a disruptive, Elon Musk-style audacity to geoengineering (also known as climate intervention) and to the small but prolific community of scientists and engineers who have spent years or even decades studying other climate-cooling methods that would operate much closer to the Earth’s surface. Members of this community would prefer that such methods never be necessary; almost in unison, they insist that geoengineering would not be a substitute for slashing greenhouse gas emissions. But they work with the intensity of people who view climate intervention as an essential hedge against disaster and want to ensure that its feasibility and risks are understood. As Douglas MacMartin, a Cornell University aerospace engineer and geoengineering researcher, explains, the goal is to “provide enough knowledge that the world . . . can make informed decisions about this” versus “knee-jerk” reactions such as “‘Oh, things are desperate. We need to go try something,’” or “‘Oh, that sounds like a bad idea. We shouldn’t do it.’”
Policy Matters
Increasingly, people in Washington agree. Citing “potentially catastrophic consequences” of global warming, a panel of the National Academies of Science, Engineering, and Medicine (NASEM) last year urged the federal government to establish, in coordination with other countries, a modest transdisciplinary research program on solar geoengineering (SG) that “attempts to moderate warming by increasing the amount of sunlight that the atmosphere reflects back to space or by reducing the trapping of outgoing thermal radiation.” The NASEM panel, which included MacMartin, said knowledge gained from the recommended research “will be critical for informing climate change response strategies, and evidence either in favor or disfavor of SG deployment could have profound value.”
As NASEM issued its findings, Congress directed the White House Office of Science and Technology Policy (OSTP) to develop a working group among the non-defense federal science agencies to manage near-term climate hazard risk and coordinate climate intervention research. The working group “should also establish a research governance framework to provide guidance on transparency, engagement, and risk management for publicly funded work in solar geoengineering research,” appropriators said.
A congressional mandate, participation of multiple federal agencies, and direction from the White House promise to elevate geoengineering from a controversial niche science to a mainstream research field relevant to policymakers. David Keith, a Harvard applied physicist and an influential proponent of geoengineering research, has been calling publicly for such a step since at least 2010. (See Prism’s October 2013 Up Close profile and February 2019 cover story.) Testifying that year before the House Science and Technology Committee, he likened geoengineering to chemotherapy as an undesirable but possibly necessary emergency measure. “We must hope for the best while laying plans to navigate the worst,” Keith told the lawmakers.
Planning for “the worst” is shaping up as prudent strategy. The world has failed to keep pace with the cuts in greenhouse gas emissions required by the 2015 Paris agreement, which aims to cap the global temperature this century at well below 2 degrees Celsius above preindustrial levels and calls for progress toward a limit of 1.5 degrees Celsius (about 2.7° Fahrenheit). The $369 billion in US climate and clean energy programs contained in the recently enacted Inflation Reduction Act falls short in fulfilling the country’s pledge to slash emissions. Some provisions do win applause from experts, however. For instance, the new law’s measures to control methane, a super-potent greenhouse gas, are “very encouraging,” says Yangyang Xu, an assistant professor of atmospheric sciences at Texas A&M University. He adds: “The direct and heavy penalty imposed on future [methane] leakage, if done with careful monitoring and verification, is a game changer, and can serve as a model for limiting other non-CO₂ emissions.”
Solar Systems
A coordinated federal research effort on geoengineering has been a long time coming. As early as 1965, a study appended to a White House environmental report found that the warming effects of a carbon dioxide buildup “could be deleterious” for humanity and urged that “countervailing climatic changes” be thoroughly explored. Led by oceanographer Roger Revelle, then director of Harvard’s Center for Population Studies, the study suggested the Earth could be cooled by increasing the albedo, or reflectivity, of the earth’s surface. One way to do that, it said, would be to spread very small reflecting particles over large areas of the ocean.
The recent NASEM panel called for government-backed research focused on three sunlight-blocking, or solar radiation modification (SRM), methods. The first, solar aerosol injection (SAI), involves discharging tiny particles into the stratosphere, an upper layer of the atmosphere between 10 and 50 kilometers (6 to 30 miles) above the Earth’s surface. The concept dates from work published in the mid-1970s by Soviet climatologist Mikhail Budyko. It gained credibility in a 2006 essay by Paul Crutzen, who shared the 1995 Nobel Prize in chemistry for discovering how pollutants in the atmosphere were destroying the ozone layer. SAI is widely considered the method most likely to work because nature has provided a proof of concept. Volcanic eruptions spew out huge amounts of ash containing sulfur dioxide. Lofted to the stratosphere, the sulfur dioxide reacts with water to form a layer of sulfuric acid droplets that reflect and diffuse incoming sunlight and radiant heat. When the Philippines’ Mount Pinatubo erupted in 1991, “stratospheric winds spread these aerosol particles around the globe,” resulting in “a measurable cooling of the Earth’s surface” for almost two years, NASA reported in 2001.
A second SRM method, marine cloud brightening, entails spraying seawater into low-lying clouds above the ocean to make them more reflective. Cirrus cloud thinning—a third, less well studied form of cooling—would break up the delicate strands of ice-crystal clouds above 20,000 feet and let heat rising from the earth’s surface escape the atmosphere. SRM methods alarm many environmentalists, primarily for two reasons: 1) they don’t remove the root cause of climate change, namely the accumulation of greenhouse gases in the atmosphere and 2) they could ease pressure on societies and governments to keep cutting emissions.
OSTP’s anticipated strategy won’t start from scratch; federal support for geoengineering research has occurred mostly under the radar but hasn’t been totally lacking. In 2020, Congress provided $4 million to the National Oceanic and Atmospheric Administration (NOAA) to initiate what the agency says is “much-needed ‘baseline’ research” on climate intervention proposals—particularly SAI. Congress upped funding this year to $9 million, instructing NOAA to expand its efforts and coordinate with NASA and the Department of Energy (DOE).
Computer modeling has provided much of what is now known about geoengineering’s potential and risks. The Intergovernmental Panel on Climate Change draws from more than two dozen modeling centers for its climate assessments, but one of the most important is the National Center for Atmospheric Research (NCAR), sponsored by the National Science Foundation and headquartered in the Rocky Mountain foothills in Boulder, Colorado. In addition to providing vital data on the Earth’s climate and weather, NCAR also serves as a nerve center for geoengineering studies, enabling global academic collaboration on experiments using ever more advanced models and a high-performance computer in Cheyenne, Wyoming.
Expect the Unexpected
The teaming up of NOAA, NASA, and DOE will expand the tools available for researchers to observe the climate system—such as satellites, balloons, aircraft capable of reaching the lower stratosphere, and, soon, a next-generation NCAR supercomputer. The anticipated result: improved models. But a host of questions remain. “Every time we go up in the atmosphere and make measurements, we find things we didn’t expect, things we didn’t know we would see,” says Gregory Frost, a NOAA supervisory research chemist.
For instance, scientists know that sulfur dioxide, injected into the stratosphere, will form aerosols and lower the Earth’s temperature temporarily—but that’s just a piece of the puzzle, notes NCAR senior scientist Simone Tilmes, a leader in solar geoengineering research. “We know we can cool, but we don’t know how much injection we actually need to cool. There’s still a huge uncertainty on how much you can cool with a certain amount of injection,” she explains. “We also need to understand the positive and negative consequences of a possible application [of SAI] and weigh risks and benefits before any of these [options] should be considered.”
As Cornell’s MacMartin puts it, “A lot of the research to date has been kind of trying things: ‘Hey, we’ll go try this strategy, scenario, climate model, and we’ll see what happens.’” Over time, the discovery of negative side effects has prompted modelers to explore ways of tweaking, for instance, the degree of cooling achieved or location of aerosol injections. Eventually, MacMartin hopes, scientists will be able to say, “We’ve looked at the response in a number of different climate models . . . Here’s what we think will happen. And here’s how confident we are in that assessment.”
MacMartin, Yale lecturer Wake Smith, and others recently studied the concept of deploying stratospheric aerosol injection only in subpolar regions. Such action wouldn’t cool the global climate, but it could halt or even reverse the melting of Arctic ice that now threatens to cause a substantial rise in sea levels, the scholars say.
Live outdoor testing of SAI would buttress existing research with accurate observations, even if the testing doesn’t answer all the outstanding questions. But an attempt by Harvard researchers last year showed just how strong public opposition can be even to research on geoengineering. The team planned to send up a balloon to release a small quantity of aerosols into the stratosphere. Following years of preparation, funded in part by Bill Gates, the team explored various launch venues. Ultimately, it partnered with the Swedish Space Corporation and made plans to use its base near Kiruna, Sweden, above the Arctic Circle. The initial flight would merely test the equipment and not spray any aerosol. But the stratospheric controlled perturbation experiment (SCoPEx) drew strong local opposition and was put on hold. The indigenous Sámi people, whose ancestral homeland stretches across Arctic regions of Sweden, Norway, Finland, and Russia, joined with Swedish environmental groups to lobby against SCoPEX, citing “risks of catastrophic consequences, including the impact of uncontrolled termination, and irreversible sociopolitical effects.”
Lists of Concerns
“I think the political barriers might be much stronger than the technical barriers,” says Alan Robock, environmental science professor at Rutgers University. Renowned for projecting the human, climatic, and ecological consequences of nuclear war—“We have to solve the problem of nuclear weapons so we have the luxury of worrying about global warming,” he says—Robock also applies his forensic research skills to geoengineering. In 2008 he published “20 Reasons Why Geoengineering May be a Bad Idea.” His tally of “risks and concerns” has since grown to 28. For SAI, they include depletion of stratospheric ozone, which helps block harmful ultraviolet rays; increased ocean acidification; a greater likelihood of droughts in some parts of the world; the need to keep increasing SAI because existing particles will grow and become less effective; and the danger of a sudden warming spike if SAI were ever shut down. “It’s like pulling back on a spring,” Robock told The Takeaway, a public radio talk show.
As concerns remain about SAI, another proposed method of solar modification, marine cloud brightening (MCB), engenders a fundamental question. That is, can you actually brighten clouds? “Clouds are a really complicated species,” says Robert Wood, a professor of atmospheric sciences at the University of Washington and principal investigator of a research collaboration on MCB. The brightening idea originated with British physicist John Latham, who proposed it two years after starting work on climate change at NCAR.
Scientists know that clouds cool the earth’s surface and believe their reflectivity can be enhanced based on observations of cloud responses to aerosols emitted in ship exhaust. “Since preindustrial times, human activity has injected a lot of aerosols and they have exerted a cooling effect on the planet that partly offsets warming by greenhouse gases,” Wood says. “So we think it’s feasible.” But, he adds, “the clouds don’t always do things that you think they’re going to do.” Their internal dynamics are too fine-grained to show up in climate models.
Testing of Latham’s theory has begun on Australia’s Great Barrier Reef, a World Heritage Site where climate change is killing coral. Wood’s team is planning tests using vessels that spray seawater into the air, forming particles of salt that would be lofted upward by warm air to low-lying clouds. Conducted over a limited area of the ocean, the tests would have minimal environmental impact but yield important information, the group says. Worldwide, however, MCB’s potential impact is unclear. According to a 2009 British modeling study, while MCB would slow the pace of global warming, it could also disrupt rainfall patterns. Some areas would likely get wetter, others drier—particularly the Amazon rain forest, “a major sink for carbon dioxide.”
‘Explore the Edges’
The notion of applying space technology to geoengineering has produced comparatively little research over the years, apart from the growing use of satellites for climate and atmospheric observations. Among those intrigued by James Early’s 1989 idea was Edward Teller, the Lawrence Livermore National Laboratory co-founder who was known as the father of the hydrogen bomb. In 1997, Teller published the idea of a space-based metallic shield to scatter sunlight. Another was Colin McInnes, now an aerospace engineering professor at the University of Glasgow, who read Early’s article as a PhD student. He went on to develop new approaches to a space shield, most recently in 2015. “It’s one of these ideas that sticks with you,” he tells Prism. McInnes went on to explore, with Cranford University colleague Joan-Pau Sánchez, a system of multiple mobile sunshades in space. This “optimal configuration” would both curb overall global warming and allow the system to adjust the sunshade effects for different latitudes and seasons, they wrote.
McInnes participated in a 2019 Harvard meeting that looked seriously at various space-based geoengineering schemes. Meeting organizers concluded that the concept “is not a plausible near-term goal or aspiration.” Still, he sees value in the research: “What I think is interesting is that [with] these concepts, you can explore, if you like, the edges of a problem or where the boundaries might be. And that then gives you a better idea of where to look for solutions.”
By not directly interfering in the Earth’s atmosphere, space-based sunshades “appear to be one of the most efficient methods to tackle climate change,” McInnes wrote in a paper coathored with Sánchez. The authors acknowledged that the project would be equivalent in scale to a Three Gorges Dam—China’s gigantic hydroelectric project—a million miles from earth and require the manufacture of reflective material equal to a decade’s worth of aluminum foil.
The 2021 NASEM report didn’t mention a space-based sunshield among its recommended federal research topics—a sign that the panel concurs with Robock’s conclusion that it’s “too expensive, too technologically questionable.” Of Space Bubbles, Harvard’s Keith says Ratti’s team “has cool tech, but when I met with them, they did not articulate any sensible reason, other than just asserting it, why this would be a better pathway than the existing pathways.” The idea generated press, he asserts, “not because it’s important but because a story with MIT and space bubbles and geoengineering was just too sweet to pass up.”
Work in Progress
Ratti is indeed a newcomer to geoengineering. Until recently, Senseable City Lab has pursued climate adaptation and mitigation “by optimizing our built environments and transportation infrastructures,” he says. The website of his Turin, Italy-based architecture and engineering firm features two examples: a synergistic pairing of autonomous taxis and a new skyscraper in Singapore, and large thermal basins floating off Helsinki’s harbor that serve as hot-water batteries for the city’s heating systems. Now, Ratti says, “Earth-based climate solutions may not be enough, and more radical technologies might be needed to address the coming climate disaster.”
While Space Bubbles is still a “working hypothesis,” Ratti’s team says it has simulated thin-film bubbles in outer space conditions and found they could prove effective at deflecting solar radiation. The spheres could be made of silicon-based melts or graphene-reinforced ionic liquids. Other potential composites will be explored. But the team’s concept paper omits details on how the bubble material reaches space, gets assembled, and is stabilized.
The challenge of filling those gaps falls to Ratti and five MIT colleagues. Two members of the National Academy of Engineering—computer scientist and roboticist Daniela Rus, winner of a 2002 MacArthur “genius grant,” and Gareth McKinley, professor of teaching innovation in the Department of Mechanical Engineering—are joined by Charles Primmerman, a Lincoln Laboratory high-energy laser expert; materials scientist Markus Buehler, a specialist in bio-inspired design and in building materials atom by atom; and aerospace engineer Paulo Lozano, director of the MIT Space Propulsion Laboratory. Ratti says that “we expect other collaborators to join us at MIT and beyond.”
The idea of lowering the Earth’s temperature has captured Rus’s attention since at least 2020, when she declared in a TEDx video: “Skeptics might tell you there is no way of reversing global warming, but then many have said we would never walk on the moon. So skepticism has never stopped scientists and engineers from doing the impossible.”
With Space Bubbles, “we aim to develop a fully reversible space-based solution,” Rus tells Prism. MIT’s Computer Science and Artificial Intelligence Laboratory, which she directs, “will develop the robotic devices and AI systems that will help control the space bubbles.”
If fabricating a vast sunshield at L1 still seems like performing the impossible, McInnes ticks off advances, like reusable vehicles, that are making space more accessible.
“If we are able to get much better at space robotics—if we can extract materials from near-earth asteroids, for example—and if we develop technologies for manufacturing large structures in space, then you can imagine a future where all of those different technological strands . . . come together” to make what now seems like an enormous technical challenge potentially more feasible, says McInnes. There’s already strong interest in in-orbit manufacturing, he adds.
Help Wanted
The biggest hurdle McInnes sees to any kind of geoengineering (space-based or otherwise)—and the reason he is a skeptic about its implementation—is governance: “the regulatory challenges of getting international agreement.” Private groups are working on the problem. They include the Carnegie Climate Governance Initiative launched by the Carnegie Council for Ethics in International Affairs, which “seeks to catalyse the creation of effective governance for climate-altering technologies” and the Global Commission on Governing Risks from Climate Overshoot, formed to “recommend a strategy to reduce risks should global warming goals be exceeded.” So far, governments haven’t been publicly involved.
Keith, an adviser to the Climate Overshoot Commission’s secretariat, says the United States and China loom large in any decision to deploy geoengineering. “If the US and China both clearly want it, then it happens. Conversely, if they both don’t want it, then it doesn’t.” If neither superpower stakes out a strong position, “it’s quite possible small countries could play a big role to determine what happens.”
If the world fails to meet the challenge of climate change and approaches catastrophe, humanity might seek a fallback in aerospace engineer Early’s science fiction plot of some 33 years ago. Venus, anyone?
Mark Matthews, Prism’s former editor, is a book author and freelance writer based in Washington, DC.