Since 1992, David Keith, a Harvard physicist and climate scientist, has audaciously proposed that we use solar geoengineering to counter climate change and the harm it does, such as extreme weather and sea-level rise. Geoengineering is defined as deliberate large-scale technological intervention in the Earth’s climate. Keith’s plan is to inject millions of tons of particles of calcium carbonate, ordinary chalk, into the Earth’s stratosphere, which extends from 10 to 50 km (6 to 30 miles) in altitude. The particles would form an aerosol, meaning they remain suspended in the air for long periods as the winds in the stratosphere spread them around the Earth. This would increase the fraction of the sun’s radiation that is reflected back into space rather than continuing to the Earth’s surface. The result would be a dimming of sunlight and a cooling trend that counteracts global warming.
This technique, called stratospheric aerosol injection (SAI) is not the only form of solar geoengineering, but it stands out as having long-term world-wide effects. That scope has raised concerns among some scientists and the public because we cannot yet predict all the consequences of SAI. The technique however has potential value, because so far international efforts have not reduced carbon emissions enough to hold the projected global temperature rise below 1.5 C (2.7 F). Scientists agree that this is a tipping point of sorts that would bring cataclysmic flooding from higher sea levels, devastating heat waves, and major harm to ecosystems.
The reduction of carbon emissions should be a central goal, but governments should also make every effort to develop solar geoengineering into a safe and predictable approach that would complement work on reducing emissions. A path to accomplishing this has just been presented in Reflecting Sunlight (2021), a report from the National Academies of Sciences, Engineering, and Medicine (NASEM), the premier scientific advisory group for the U. S. government. Produced by a panel of distinguished scientists and technologists, and experts in public policy, ethics, and conservation, the report recommends that the federal government substantially fund essential research in the science, technology, and social impact of solar geoengineering.
This would build on the established fact that aerosols in the stratosphere provide cooling that could counter the temperature rise, as has been observed in volcanic eruptions. In 1816, “The Year Without a Summer,” summer temperatures in Europe were the coldest in 200 years, resulting in massive crop failures and food shortages. The cause was the eruption in April 1815 of Mount Tambora in the Dutch East Indies (now Indonesia), which ejected megatons of ash, and the volcanic byproduct sulfur dioxide (SO2) which combined with water to form droplets of sulfuric acid. With the ash, these formed an aerosol that reduced the Earth’s average land temperature by 0.4 C – 0.7 C (0.7 F- 1.3 F). Then in 1991, Mount Pinatubo in the Philippines erupted and spewed an estimated 20 megatons of SO2 and ash 20 km (12 miles) into the stratosphere, reducing global temperatures by 0.6 C (1.1 F) between 1991 and 1993.
A cooling of less than 1 degree C might seem small, but it is comparable to 1.1 C, the rise in global temperature since pre-industrial days, which is already causing problems; and to 1.5 C, the point where these problems would grow exponentially. In fact, data from volcanic eruptions and computer modeling show that SAI could keep warming below 1.5 C.
For example, in 2018, the International Panel on Climate Change (IPCC, the U. N.-sponsored scientific group that tracks climate change), used estimates by Keith and others to conclude that high-altitude aircraft could deposit enough SO2 aerosol to offset a temperature increase of 1.5 C. The yearly cost would be $1 billion to $10 billion, far less than the trillions it would cost to revamp industry to reduce carbon emissions. In 2021 Cornell climate scientist Douglas MacMartin similarly calculated that injecting ten megatons of SO2 into the stratosphere annually (compared to an annual 100 megatons of SO2 from industrial pollution) would keep warming below 1.5 C. This, he noted, could “avoid catastrophic sea-level rise, and limit the risk of forest fires and hurricanes.”
These results seemingly point to a relatively direct and affordable solution to what has so far been the intractable problem of climate change, so why do we not simply go ahead and start working on SAI? The answer is that we must be cautious before applying SAI because of the sheer size and complexity of our planet’s climate and the possibility that intervention would do more harm than good.
The huge scale of climate processes has been a factor since the early days of studying human-induced climate change. In 1938 the English engineer Guy Callendar published the first detailed study of the rise in the Earth’s temperature over past decades and made the case that this came from human activities that increased the concentration of atmospheric CO2. His paper was either ignored or criticized at the time, although it is now a classic. Much of the negative reaction reflected the powerful but unsubstantiated belief that human actions could not possibly affect an entire planet. Today that unintended human impact is mostly taken for granted, but now we must ask a new question about the human role: do we know enough about the vast natural processes that determine the Earth’s climate and temperature to dare tamper with them?
This question has produced strong opinions in the climate science community. In 2015, for instance, geophysicist Raymond Pierrehumbert, then at the University of Chicago and now at Oxford University, wrote a searing article attacking SAI. He called it “wildly, utterly, howlingly barking mad” on the grounds that it involves large risks, both known and unknown, that should put it completely off the table when we consider that “we have only one planet to live on.” Pierrehumbert was also concerned that even if SAI works to reduce temperatures, it is only a short-term Band-Aid; yet it could displace efforts toward expediting the true long-term solution, reducing the amount of CO2 in the atmosphere, whereas SAI would not affect this root cause of our climate problems. Others have expressed similar concerns, although less vehemently.
But no one is proposing that solar geoengineering replace the essential efforts to reduce atmospheric CO2, only that it should complement them. That increasingly looks necessary because so far, the world lacks the will to adequately reduce emissions. In November 2021, 120 nations represented at the twenty-sixth annual U. N.-sponsored climate conference could not agree to reduce emissions so as to limit global warming to 1.5 C. The latest IPCC report (April 4, 2022) concludes that major “societal transformation” is necessary to reach that goal, but that the time frame to do so is very short and immediate action is needed – if the international community even decides to take action. Facts like these motivate Keith to propose that solar geoengineering supplements work on reducing CO2 so that together they limit the temperature rise to 1.5 C, provided that properly designed and governed research addresses the real or perceived problems with geoengineering.
Reflecting Sunlight brings the weight of its panel deliberations to its recommendation of a similar program. The report concludes that solar geoengineering could indeed help limit global warming, but also that there are “many uncertainties and possibilities for unintended harmful consequences.” This makes it essential to answer both the technical question, does solar geoengineering work; and the technical-societal question, does it work safely, which requires that we consider the related “societal perceptions and reactions, political and economic ramifications, and ethical concerns.” The panel emphatically supports deeper investigation by recommending that for the first time, the federal government fund research in solar geoengineering with an investment of $100 million to $200 million over five years. Twenty percent of this should support research governance, such as establishing procedures for public transparency, for decision-making about research directions and their consequences, and for evaluating other related moral issues.
Reflecting Sunlight is notable in formally recognizing solar engineering as potentially valid for dealing with climate change, but it also inspired a backlash. One published critique of the report objected to it as “mainstreaming” solar geoengineering and encouraging its unilateral deployment, calling instead for the U. S. to join with other countries to “phase out fossil fuels, advance global climate action, and invest in climate justice.” A later response rebutted much of the critique. This back-and-forth is one illustration that the social-political issues raised by climate engineering remain unresolved.
A proposal for the first field test related to SAI also illustrates that societal perceptions of solar geoengineering matter. Frank Keutsch, an atmospheric scientist at Harvard, is the principal investigator and David Keith is also involved with the Stratospheric Controlled Perturbation Experiment (SCoPEx). It is not supported by the federal government but by internal Harvard sources, which are partly funded by outside organizations and individuals including Microsoft’s Bill Gates. The project’s aim is not to immediately initiate full-scale SAI, but to better understand the dynamics, physics and chemistry of aerosol particles in the stratosphere to improve predictions about the behavior of SAI.
The plan for initial testing is to use a high-altitude balloon to carry instrumentation to a height of 20 km (12 miles), well into the stratosphere. There the balloon will release a 0.1 kg to 2 kg (0.2 to 4 pound) sample of chalk particles, too small to alter the climate but producing an aerosol plume that would be probed by the instruments. The first step toward this limited research was to have been to launch the balloon and its instruments in a test run without deploying any aerosol material. The researchers had arranged to do this in the summer of 2021 in northern Sweden, with the help of a Swedish space company.
But in early 2021, Swedish environmental groups and a council representing the indigenous Saami people demanded that the test be canceled because the project “entails risks of catastrophic consequences.” After some delay, SCoPEx announced that the test had been put on indefinite hold. Keith’s response was “I was frustrated…My strong impression is that there is growing agreement in the international and environmental community that research on this topic makes sense.” However, the critique of Reflecting Sunlight and the intense reaction against the test in Sweden (even though the test would not release anything into the stratosphere) show that SAI and its preliminary research would not be easily accepted. That is why it is significant that Reflecting Sunlight recommends funding for public outreach and thoughtful consideration of consequences.
Equally important, Reflecting Sunlight considers two alternatives to SAI that are not of world-wide scope and so present fewer actual or perceived risks. The first, marine cloud brightening, would provide cooling by creating reflective clouds at altitudes from sea level to 3 km (2 miles) over selected parts of the ocean. We know that ocean-going ships display “ship tracks” caused by aerosols emitted as pollution from the fossil fuel the ships burn. These aerosols act as nuclei where water condenses into droplets to form clouds, and the more droplets, the more reflective the cloud. With this as a model, the idea in marine cloud brightening is to spray a fine mist of salt water into the atmosphere over the ocean to produce reflective clouds.
Rather than being widely applied to cool the whole Earth, this technique would operate on a smaller regional scale, potentially allowing fine-tuning such as protecting a particular coral reef system. However in a recent email interview, Sarah Doherty, who studies cloud brightening at the University of Washington, noted that current predictions about the effectiveness and any side effects of the method rely largely on computer modeling. Her research group and others are now contemplating field studies at a scale too small to alter climate, to gather new data that will complement and improve the computer analysis.
The second method, cirrus cloud thinning, would cool the Earth differently than by dimming the sun. Cirrus clouds are made of ice crystals at high altitudes. They reflect some sunlight, but are more effective at blocking heat in the form of infrared radiation rising from the Earth. Thinning the clouds would allow more heat to escape into space with subsequent cooling. The proposal is to seed the clouds with ice nuclei such as mineral dust, creating bigger but fewer ice crystals. This would thin the clouds, and the larger crystals would fall faster, reducing the lifetime of the clouds. However, Ulrike Lohmann, an atmospheric scientist at ETH Zürich in Switzerland who specializes in cirrus clouds and their possibilities for geoengineering, cautions that “For the time being, cirrus cloud thinning should be viewed as a thought experiment” that so far has been explored only with computer simulations.
Reflecting Sunlight and reactions from scientists and the public show that the answer to the question “Should we geoengineer sunlight?” is “Not yet, but we must give this approach serious consideration.” If new research shows that it can be used safely and effectively, we should deploy it together with international plans to reduce emissions. One active geoengineering researcher, Sarah Doherty at the University of Washington, concisely sums up this hybrid approach. “As scientists,” she wrote in an email, “we hope to help decision-makers conclude whether it is better to let climate warming play out, based on the available policy toward greenhouse gases, or if climate intervention could reduce climate risk while we return to a world with safer levels of greenhouse gases.” There are questions about the safety and efficacy of solar geoengineering, but there is absolutely no question that left unchecked, climate warming will wreak further harm on our planet and ourselves. It would be a grave mistake to overlook any technology that could help us mitigate these threats. Clearly, governments and policy-makers ought to follow the recommendations in Reflecting Sunlight, and begin programs to determine whether and how solar geoengineering should be included among the tools we use against this looming catastrophe.