This story is over 5 years old.


Harvard Scientists Say It's Time to Think About Engineering the Climate

Wait, isn't that the plot of 'Snowpiercer'?
​Image: ​NASA

There's ​a black sheep Plan B that some top tier scientists have been investigating to fight global warming, and in order to learn enough about it—or rule it out completely—they say we might just have to float a giant balloon with a solar-powered monitoring system into the stratosphere.

On November 17, 2014, Harvard's School of Engineering and Applied Sciences published a research announcement headlined "Adjusting Earth's Thermostat, With Caution." That might read as oxymoronic—intentionally altering the planet's climate has rarely been considered a cautious enterprise—but it fairly accurately reflects the thrust of three new studies published by the Royal Society, all focused on exploring the controversial field of geoengineering. The first of the bunch features a new proposal for real-world experiments from four distinguished Harvard scientists.


Geoengineering, of course, describes a variety of large-scale efforts to adjust the planet's climate to counteract global warming. If policymakers and citizens fail to reduce emissions, the reasoning goes, a planetary-scale technical fix may buy us some time to avert disaster.

The papers, each of which were co-authored by the renowned physicist and geoengineering research proponent ​David ​Keith, primarily examine one particular mode of climate tinkering—solar radiation management. Often shortened to SRM by science wonks, it most typically consists of dispersing sulfate aerosols—sulfuric acid—into the atmosphere, where they would bounce back a small percentage of incoming sunlight, thus cooling the planet.

The effect is not unlike a volcanic eruption, which disperses sun-blocking particles into the sky—which is exactly why it's the most-studied geoengineering proposal out there. (It's no coincidence that ​another new study published this week found that small volcanic eruptions help slow the rate of global warming). It's also relatively cheap: For a few billion dollars a year, one could theoretically cool the globe a matter of degrees.

That's why Keith and his colleagues want to experiment.

"Controlled experiments allow quantitative confrontation between experiments and theory," Keith wrote me in an email. "That is how we learn things in science. We should expect surprise. Our first goal is to improve understanding of the amount of ozone loss that might be caused by stratospheric aerosol geoengineering. The experiment also allows us to test aerosol dynamics involved in the production of fine particles in the stratosphere."


The ​first paper in the batch, co-authored with fellow Harvard scientists John G. Anderson, Debra Weisenstein, and John A. Dykema, outlines a proposal for conducting a small-scale experiment—floating the aforementioned monitoring apparatus into the stratosphere on a balloon—to examine the chemical effects of dispersing aerosol in the atmosphere. There are two models for doing so, the more ambitious of which includes a craft that could move at 8 meters per second through the sky, seeding aerosol and taking measurements. The other would putt along at a meter per second.

This "stage two system architecture" of the concept suggests a system that would lift a vehicle called the StratoCruiser into the, yeah, stratosphere, with a giant balloon, to gather info on releasing sulfate aerosols into the yonder.

The idea is to discover, among other things, whether those aerosols would harm the ozone. They would use a very tiny amount, and stress as much in the paper, in part to combat any potential criticism from environmentalists that they're intentionally polluting the sky. The paper notes that "less than 1 kg of sulfuric acid is needed per flight, an amount that is less than the amount of sulfur released by one commercial passenger jet in 1 min of flight time."

In other words, it's about as bad for the environment as one minutes' worth of your flight home for Christmas.

But the symbolic concerns loom larger—the very notion of tampering with the atmosphere makes people queasy, justifiably (reflected not least in the long-running group of conspiracy theorists ​dedicated exclusively to chemtrails). There's also the fear that once people get the notion there's a technical fix for climate change out there, they won't muster the will to beat climate change the old fashioned (and most important) way—by weaning our economies of coal and oil.


"We're having a public debate—or a battle, even—in the academic community," Keith told me in a phone interview. "Public willingness is kind of ambiguous," he says, which might be because the research hasn't largely penetrated the mainstream. But "lots of our academic colleagues don't want there to be experiments."

That's true. I saw those fault lines laid bare at the first international climate engineering conference (CEC 2014) in Berlin this year. Even at a meeting dedicated to the topic, the opponents of experimentation at times seemed to outnumber the proponents.

Keith preempted this sentiment in the release for the research itself: "The idea of conducting experiments to alter atmospheric processes is justifiably controversial, and our experiment, SCoPEx [stratospheric controlled perturbation experiment], is just a proposal." In fact, Keith and fellow Harvard professor James G. Anderson have been proposing a version of this for years now.​The New York Times called it a "tiny geoengineering experiment." It hasn't taken off in part because of the persistent controversy in the academic community—those voices warning of opening pandora's climate box—and subsequent lack of public funding.

"Obviously I'm on the side of science. When there's something that's potentially useful to the world," he told me, "turning away from learning more is really odd. We're trying to make the case that there are a whole fleet of experiments that could teach you meaningful things about the risks and benefits of climate engineering."


Keith isn't the only respected scientist who wants to see more emphasis on engineering in research circles. Earlier this year, Stanford atmospheric scientist Ken Caldeira, ​perhaps the top modeler of climate engine​ering, told me at CEC 2014, "For the current generation of climate models, and the way people measure climate damage… at modest levels of solar geoengineering, everyone is better off than without it."

It's a little surprising that Harvard's announcement didn't make more of a stir—we're talking about climate engineering here; artificially attempting to regulate the globe's temperature is a touchy subject, and has, before this era of legitimate scientific inquiry, mostly been the fiction-laced province of hucksters and supervillains. Bona fide scientists haven't been wading into the climate-alteration game until relatively recently.

The last time they attempted to, at least in any seriously visible way, ended in a bit of PR kerfuffle. The Stratospheric Particle Injection for Climate Engineering (SPICE) project, a collaboration between the Universities of Cambridge, Bristol, and Oxford, intended to float a balloon into the sky and have it spray out regular old water droplets, in order to test out the hardware required for future geoengineering experiments. It was cancelled after a potential conflict of interest arose over the researchers having apparently filed patents for technology similar to the gear involved—and, at least in the press narrative, due to opposition from environmental nonprofits.


"The SPICE experiment is designed to develop cheap tools for deployment," Keith wrote me in an email. "The one problem we do not have with this technology is it's cost. It was recklessly premature to focus on cheap deployment. In my view the only justifiable goals for early experiments is to improve understanding of risks or efficacy. Our experiment does both."

Keith and his colleagues are hoping to avoid a dustup like that, in part, by carefully examining the ethics and setup of geoengineering experiments. The second paper published this week, Field experiments on solar geoengineering: report of a workshop exploring a representative research portfolio, describes exactly that. It's a pretty meticulous examination of most of the major proposed geoengineering techniques, and of what would have to happen before they could ethically and feasibly become the subject of experiments in the field.

He tells me he'd like to see a review panel set up to examine, approve, and help authorize potential geoengineering experiments. Right now, "program managers don't even want to meet because they're so afraid of this," he says.

In order for that to happen, the National Academy of Science has to finish its ​technical evaluation of geoengineering experiments, which is expected to be published in a few months. After that, Keith says, is the earliest we could see bona fide experiments get a greenlight. "I would say the earliest announcements of opportunity months after the NAS report is out," he tells me. Which means we could see small-scale, controlled geoengineering experiments by "late summer or fall."


It could be longer. The biggest controversy of CEC 2014 was ​an effort to establish a set of guidelines for future climate engineering experiments—at the packed town hall meeting where scientists both for and against it gathered, the proceedings could best be described as polite pandemonium. Protests, from inside academia and out, may prolong real-world experimentation.

Many people assume that solar geoengineering would be used to suddenly restore the Earth's climate to preindustrial temperatures

But if we do end up experimenting with climate engineering, we're going to have to think about when and how it would—if ever—be appropriate to deploy. To that end, ​the final paper in the Royal Society batch discusses a "new approach" to climate engineering.

"Many people assume that solar geoengineering would be used to suddenly restore the Earth's climate to preindustrial temperatures," Keith said, "but it's very unlikely that it would make any policy sense to try to do so."

Instead, he and fellow scientists Douglas G. MacMartin and Ken Caldeira propose a framework for "finite" geoengineering—essentially, it entails analyzing how much temperature rise we should be seeing thanks to humanity's contributions to amplifying the greenhouse effect, and temporarily erasing the extra swelter with an annual aerosol injection, while drawing down carbon emissions on Earth.

"We thus describe an alternative scenario in which solar geoengineering is used only to constrain the rate of change of global mean temperature," the authors write, "this leads to a finite deployment period for any emissions pathway that stabilizes global mean temperature." It's geoengineering as a limited countermeasure. In the past, Keith has advocated sending a fleet of jets into the sky to disperse the aerosols, as the most cost-effective way to carry out such a plan.

Here are the rough calculations: If humankind maxes out our rate of greenhouse gas emissions to the point that it leads to a rise of 0.1°C per decade, the scientists figure it would require annual geoengineering—dusting the atmosphere with a load of sulfate—for 160 years to keep temperatures in the 'normal' human-friendly range. If we head toward an even heavier emissions scenario before pulling back, it could be twice that long. That's 320 years of geoengineering, folks—three centuries of jets dusting the planet with faux-volcanic spray.

Meanwhile, Keith says he wishes we were all doing more to reduce emissions, and that he does everything he can in his personal life to do the same. He tells me that there really hasn't been any big surprises in the climate science for ten years, and that the Intergovernmental Panel for Climate Change, which is responsible for delivering climate policy recommendations to governments, has become "too bureaucratic." He used to be an author.

Now, much of his time seems devoted to getting other scientists to even consider more climate engineering research. We should be thinking about this, is his point.

"I guess I'm one of the very few scientists saying this," Keith told me. "If it holds up, the benefits of doing a moderate amount of geoengineering are really large in the global and the risks are really small."