Roundtables
ShareHas the time come for geoengineering?
Scientists have long studied and debated the promises and perils of deliberately influencing Earth's weather and climate systems. But today, faced with ever more pessimistic predictions about the pace of global warming and the irrevocable damage it could do to the planet, some are talking seriously about implementing theoretical geoengineering schemes such as blocking the sun as an emergency response. In "20 Reasons Why Geoengineering May Be a Bad Idea" (May/June 2008 Bulletin), Alan Robock raises a host of scientific, social, and ethical issues posed by geoengineering. Below, Robock and his four fellow discussants debate how to weigh geoengineering’s potential benefits against its negative consequences.
Although Alan Robock's "20 Reasons Why Geoengineering May Be a Bad Idea" raises legitimate questions, it seems to argue against implementation rather than against studying the underlying science. Few people are actively advocating for immediate, full-scale implementation of geoengineering techniques as a means of addressing climate change. But many people are suggesting that we learn more about the efficacy of such techniques--including Alan, who was recently awarded a National Science Foundation grant to study the effectiveness and possible consequences of injecting aerosol particles into the stratosphere to reduce incoming solar radiation.
In terms of geoengineering concerns, it's helpful to group them into three categories:
Efficacy. Clearly, any geoengineering technique first needs to achieve the intended effect for a reasonable cost--whether the goal is buying time for more sustainable solutions by reducing incoming solar radiation or addressing the root cause of warming by removing carbon dioxide from the atmosphere. It's critical that scientists be allowed to study efficacy through experimentation and modeling without being stigmatized by the assumption that their work will cause a rush to full implementation.
Impact. The environmental impacts of the technique must either be minimal or acceptable relative to the benefits of action and the consequences of inaction. Martin Bunzl makes this point clearly in "An Ethical Assessment of Geoengineering," an accompanying essay on p. 18 of Alan's article. In the case of ocean iron fertilization, 12 small, open-ocean experiments have already been conducted by oceanographers to improve understanding of both efficacy and impacts. (See "Mesoscale Iron Enrichment Experiments 1993-2005: Synthesis and Future Directions".) From the start, it was clear that these experiments weren't a danger to the environment and that their effects wouldn't last long. Scientists should be encouraged to study impacts through experimentation and modeling as long as it can be reasonably presumed that their impacts are short-lived.
Implementation. If a technique is both effective and sensitive to the environment, the following implementation questions become important: Who implements it? Who regulates it? And how do we incorporate these activities into existing regulatory and legal frameworks and treaties? These questions are difficult but not intractable, as many carefully negotiated international agreements already demonstrate, including the International Maritime Organization's London Convention on ocean dumping (signed by 80 countries in 1972, including most of the developed world) and the U.N. Law of the Sea treaty (signed and ratified by most countries except the United States).
More broadly, the provocative title of Alan's article and the quick treatment of individual concerns obscure the complexity behind these subject areas--as Martin and Ken Caldeira have addressed. We expected a summary of research results suggesting a priori that geoengineering is a bad idea, but didn't find one. Also, we found it distracting that many of Alan's concerns (i.e., ozone depletion, acid deposition, effects on cirrus clouds and plants) are specific to one technique--aerosol seeding--but offered as reasons why geoengineering in general is a bad idea. Another of Alan's examples presumes that "humans [adopt] geoengineering as a solution to global warming, with no restriction on continued carbon emissions." No one is suggesting that geoengineering replace emissions reduction.
Most surprising is Alan's conclusion that global warming is a not a difficult technical--but rather purely a political--problem, and therefore, geoengineering isn't required to solve it. We disagree. The road ahead is paved with difficult technical challenges in addition to the considerable political ones. Many new, clean technologies that promise incremental improvements in efficiency also require substantial scientific achievements--such as genetic modification of organisms to make novel substances (i.e., enzymes that process various feedstocks for cellulosic ethanol) or revolutionary advances in materials and process sciences (i.e., new thin-film technologies for solar power). Emission reductions don't simply follow from mandates; we must innovate alternatives to fossil fuels.
That we need to contemplate geoengineering to buy us time for that innovation is unfortunate. That we have the scientific, technical, and human potential to do so responsibly is not.
Winston Churchill once famously said, "Democracy is the worst form of government except all the others that have been tried." Climate engineering may indeed be a bad idea, but so far, better ideas to mitigate global warming show little traction.
We can all agree that eliminating carbon emissions is the right thing to do and that everyone in the world should set aside narrow short-tem self interest and instead work together to provide a better environment for future generations. And when Alan Robock provides 20 reasons why geoengineering may be a bad idea, we can all agree--it is a bad idea. However, what we appear to be achieving in the meantime may be much worse. For all the recent talk about reducing carbon dioxide emissions, the concentration of atmospheric carbon dioxide is growing more rapidly than supposedly pessimistic scenarios predicted even a few years ago.
Preliminary climate model simulations show that in a high-carbon-dioxide world, Earth's climate would be more similar to that of several centuries ago with climate engineering than without it. It won't work perfectly, but imperfection isn't an argument against improvement. The question is whether, in the face of rising greenhouse gas concentrations, climate engineering will improve environmental conditions or merely make things worse. This is an open research question that needs to be vigorously pursued, but an examination of a few of the criticisms on Alan's list demonstrates that so far, none are nonnegotiable:
Continued ocean acidification. Emissions reduction and climate engineering are two levers of action that can be employed jointly or separately. Ocean acidification is a consequence of excess atmospheric carbon dioxide getting dissolved into the ocean, not climate engineering. Climate engineering cannot reverse every adverse consequence of carbon dioxide emissions, but no thoughtful person ever claimed it would.
Ozone depletion. The Mount Pinatubo eruption lofted more than enough aerosols into the atmosphere to compensate for a doubling of atmospheric carbon dioxide, yet ozone concentrations fell by only 3 percent. And it's believed that this small reduction was caused by chlorine from human-made chlorofluorocarbons, which are now banned by the Montreal Protocol. So while the threat to the ozone layer is worth studying in greater detail, it's expected to diminish with time. Furthermore, schemes have been proposed that might preferentially scatter ultraviolet radiation, compensating for any minor reduction in the protection that the ozone provides us from ultraviolet light.
Effects on plants. Alan is correct that we need to study possible effects of climate engineering on plant growth. After the Mount Pinatubo eruption, vegetation everywhere grew more vigorously, taking up more carbon from the atmosphere. This is because diffuse sunlight is able to reach down to enhance photosynthesis in the lower leaves of forest trees, which are normally shaded by the upper canopy in direct sunlight. In general, plant growth responds almost linearly to changes in the amount of sunlight--a 2-percent reduction in sunlight might be expected to produce 2 percent less photosynthesis. But people growing crops in greenhouses often elevate the carbon dioxide level to fertilize their plants, and this effect is typically larger than 2 percent. Therefore, it's possible that a high-carbon-dioxide world with slightly reduced but more scattered sunlight would have higher crop yields than today's world. In computer simulations, vegetation grew more vigorously in an engineered high-carbon-dioxide world than it did in the natural low-carbon-dioxide world. Of course, we can expect these changes to affect natural ecosystems in unforeseen ways, and so should certainly be the subject of intense study.
More acid deposition. The amount of sulfur used in a climate engineering system would be a small percentage of today's emissions from power plants. So if current sulfur emissions regulations were tightened by a few percent when such a system was deployed, there would be no increase in overall sulfur emissions. Furthermore, there's nothing magical about sulfur--other compounds such as silica or calcium carbonate could be used to scatter incoming sunlight, although perhaps at somewhat greater economic cost.
I'll leave Alan's other points for future missives, but the take-home message is, preliminary climate model simulations indicate that climate engineering may mitigate some but not all of the effects of rising greenhouse gas concentrations. While we might prefer near-universal cooperation in carbon dioxide emissions reduction, it's clearly time to plan what we will do if those emissions reductions don't come quick enough or are not deep enough to prevent a climate crisis. The question isn't whether we need to plan for such an eventuality, but what form that planning should take.