Three different classes of geoengineering identified by the American Meteorological Society and the Royal Society have very different risks and time scales and would play very different roles in a climate strategy.
Climate Remediation Technologies
Climate remediation is similar in concept to cleaning up contamination in our water or soil. The first problem is to stop polluting (mitigation) and the second is to remove the previously emitted contaminants (remediation) and put them somewhere—for example, filter CO2 out of the air and pump it underground.
Climate remediation technologies are, with some exceptions, relatively safe and noncontroversial and have relatively few governance issues. They address the root cause of the problem, but these methods work slowly. It would take years if not decades to reduce the concentration of CO2 in the atmosphere through air capture and sequestration. These technologies are also expensive when compared to the option of not emitting CO2 in the first place. However, as we try to bring emissions close to zero, it will likely remain difficult to operate heavy-duty transportation without liquid hydrocarbons. If the sustainable supply of biofuels can’t match this demand, a choice may be to continue the use of fossil fuels and offset the resulting emissions by removing CO2 from the atmosphere. As well, we may decide that the atmospheric concentration must be brought down below stabilized levels, perhaps even below 350 parts per million (ppm). If we don’t want to wait many hundreds of years for this to happen through natural processes, we may have to actively remove greenhouse gases from the atmosphere.
Air capture technologies are closely related to carbon capture and storage (CCS) technologies. For CCS, we are contemplating separating out CO2 after coal combustion and then pumping it deep underground into abandoned oil or gas fields or saline aquifers. The technologies for removing CO2 from the air (air capture) and from flue gas are similar. After capturing the CO2, it has to be put somewhere isolated from the atmosphere. Currently, we are considering geologic disposal: pumping the CO2 deep underground. The implementer must obtain rights to the underground pore space and be able to assign liability for accidents, leakage, and so on. These same issues exist for storage of CO2 in a CCS project. However, Keeling has suggested that the amount of CO2 we may need to remove from the atmosphere is such that we will have to consider disposal in the deep ocean as a form of environmental triage.1 Ocean sequestration would clearly involve much more serious governance issues.
Beyond air capture, the Royal Society report on geoengineering lists a number of other carbon-removal technologies, including augmentation of natural geologic weathering processes and biological methods such as reforestation.2 Of these, biological methods, which include genetically modified organisms (GMOs) and ocean iron fertilization, would have governance issues similar to climate intervention discussed below.
The purpose of climate intervention is to modify the energy balance of the atmosphere in order to restore a prior radiation balance. Climate intervention has also been called solar radiation management (SRM), or sunblock technology, and some consider these technologies to be a radical form of adaptation. If we can’t find a way to live with the altered climate, we intervene to roll back the impact.
Volcanic eruptions, which emit massive amounts of sulfates that reflect sunlight, cause colder temperatures for months afterward. Such rapid global temperature decline can be simulated in climate models by changing the global heat balance. This is evidence that climate interventions that change the radiation balance of the Earth could be effective at reducing global temperatures.
Climate intervention techniques to change the radiation balance are amazingly inexpensive, especially compared to mitigation, and all are relatively fast acting. For example, just three grams of sulfur aerosols can offset the warming of a ton of CO2.3 Some methods could lower temperatures within months of implementation, but they do not “solve” the problem in that they do nothing to reduce the excess greenhouse gases in the atmosphere. So, if we reflect more sunlight and don’t reduce CO2 in the atmosphere, the oceans will continue to acidify, severely stressing the ocean ecosystems that support life on Earth. And if we keep adding CO2 to the atmosphere, we will eventually overwhelm our capacity to do anything about it with geoengineering. So, climate intervention can’t be a standalone solution. It is, at best, only a part of an overall strategy to reduce atmospheric concentrations of greenhouse gases and adapt to the unavoidable climate change coming down the pike.
There are ideas for putting reflectors in space and increasing the reflectance of the oceans, land, or atmosphere. Some propose global interventions such as injection of aerosols (sulfate particles or engineered particles) in the stratosphere, and the Novim report spells out the required technical research for this concept in some detail.4 Others propose more regional or local interventions, such as injecting aerosols in the Arctic atmosphere only in the summer to prevent the ice from melting.5 Even more local, and perhaps most benign, is the idea of painting rooftops and roadways white to reflect heat.
Climate intervention has some serious drawbacks. It may be difficult to predict exactly how the weather patterns will change. The intervention may cause the climate in some parts of the Earth to improve and in others to become worse. It will be very difficult to determine whether these deleterious conditions arise simply from climate variability or are due to the intentional intervention. Climate model simulations have shown that if we were to suddenly stop a global intervention, the global mean temperature would quickly return to the trajectory it was following before the intervention. This means that temperatures could increase very rapidly upon cessation of an intervention, which would likely be devastating. Ironically, then, climate intervention may only provide temporary respite but would be difficult to stop. These drawbacks mean that climate interventions are unlikely to be deployed until or unless we have strong evidence that the risks of climate change plus climate intervention are less than the risks of climate change alone.
Climate intervention might nevertheless be an effective part of an overall climate strategy. We already emit millions of tons of aerosols in the form of air pollution, which is estimated by the Intergovernmental Panel on Climate Change (IPCC) to be masking roughly one half of the global warming that would otherwise be caused by today’s atmospheric greenhouse gas concentrations.6 So, as we clean up air pollution to protect human health, or stop emitting air pollution as we shut down coal-fired electricity generation in mitigation efforts, we will also cause a significant increase in short-term warming. (Long-term warming remains largely a function of the CO2 concentration.) We may want to offset this additional warming by injecting aerosols into the stratosphere, where they are even more effective at reflecting radiation. This plan might cause much less acid rain and improve human health impacts, compared to power plant and automobile emissions, while continuing to mask undesirable warming. It is possible that the “drug” of aerosol injection could be a type of “methadone” as we withdraw from fossil fuels.
The “Catch-All” Category
The catch-all category includes technologies to manage heat flows in the ocean or actions to prevent a massive release of methane in the melting Arctic. These technologies are not as well understood or developed, but the classification recognizes that not all the ideas are in. Also, we may need to address some very specific global- or subglobal-scale emergencies caused by climate change.
For example, recent studies have shown vast amounts of methane, a powerful greenhouse gas, are leaking from the Arctic Ocean floor. Abrupt increases in methane emissions have been implicated in mass extinctions observed in the geologic record and could trigger runaway climate change. (It is the possibility of such runaway climate change that most clearly supports the need for geoengineering research.) James Cascio recently posed an idea for deploying genetically engineered methanotrophic bacteria (bacteria that eat methane) at the East Siberian Arctic Shelf. Is this possible? What are the risks? Could the release of genetically modified methanotrophic organisms cause problems in the Arctic ecosystems? This may be an idea with merit—or it may be a very stupid idea. A geoengineering research program should include funding to freely explore theoretical ideas and perform the modeling and laboratory studies to distinguish between concepts that are worthy of more work and those that are completely impractical or too dangerous.
There should also be a “top-down” research program that examines potential emergencies that could result from climate change and then attempts to design interventions for these specific situations. Although higher temperatures will be a serious problem, other impacts of climate change might be more critical. Volcanic eruptions have effects that are similar to the effects of aerosol injection. Although these reduce temperature, they also reduce precipitation. Reducing precipitation would clearly be a bad thing to do. By looking only at what we know how to do (reduce temperatures) versus what problem we want to solve (increase water supply), we could be making conditions worse. Geoengineering research should not just be structured around the “hammers” we know about. We should also collect the most important “nails” and see if we can design the right hammer.
Thus, we might try to develop methods that directly attack specific, important climate impacts. Can we conceive of a way to control the onset, intensity, or duration of monsoons to ensure successful crops in India? Can we conceive of a way to stop methane burps or hold back melting glaciers? Some part of a geoengineering research program should take stock of the possible climate emergencies and then look for ideas that would ameliorate these problems.
None of these geoengineering technologies should be considered a standalone solution; some or all of these could be integrated into a comprehensive climate-change strategy that starts with mitigation. Such a comprehensive strategy might include:
- A steady, but aggressive transformation of the global energy system to eliminate emissions with concurrent elimination of air pollution in a few decades (mitigation).
- Carbon removal over perhaps 50 to 100 years to return to a safer greenhouse gas concentration (climate remediation).
- Time-limited climate intervention to counteract prior emissions and reductions in air pollution, tapering off until greenhouse gases fall to a “safe” level (climate intervention).
- Specific, focused actions—such as preventing methane burps or melting Arctic ice—to reverse regional climate impacts (technologies from the “catch-all” category).
- Keeling, RF. Triage in the greenhouse. Nature Geoscience 2, 820-822 (2009).
- Shepard, J et al. Geoengineering the Climate: Science, Governance and Uncertainty (London, Royal Society, 2009).
- Keith, D. Geoengineering research. Presentation to National Academy of Science America’s Climate Choices conference (June 2009) [online].
- Blackstock, JJ et al. Climate Engineering Responses to Climate Emergencies (Novim, Santa Barbara, CA, 2009) [online]. arxiv.org/pdf/0907.5140. people.ucalgary.ca/~keith/Misc/Keith_NASJune2009.pdf.
- MacCracken, M. On the possible use of geoengineering to moderate specific climate change impacts. Environmental Research Letters (2009).
- Climate Change 2007: The Physical Science Basis. IPCC Fourth Assessment Report (Cambridge University Press, Cambridge, UK, 2007)