CSTPR has closed May 31, 2020: Therefore, this webpage will no longer be updated. Individual projects are or may still be ongoing however. Please contact CIRES should you have any questions.
Ogmius Newsletter

Ogmius Exchange


Photo of Bill TravisThis edition of Ogmius features an article by Center director Bill Travis discussing the pros and cons of framing geo-engineering as an emergency response to global warming.  Bill took part in a panel discussion titled "From Research to Field Testing and Deployment: Ethical Issues Raised By Geo-engineering."  The panel was part of the National Academies’ workshop "Geo-engineering Options to Respond to Climate Change: Steps to Establish a Research Agenda."  Bill argued at that workshop that experience with weather modification indicates a public willingness to accept interventions like cloud seeding for water resources, but a strong public aversion to larger-scale interventions like hurricane modification.  Comments welcome!  admin@sciencepolicy.colorado.edu

Emergency Use Only: Geo-engineering to Reduce Global Warming
by William Travis

Photo of geoengineering plantThe geo-engineering response to the threat of global warming is often framed as an emergency measure---a back-up, plan-B, last-ditch response. This notion is expressed in, for example, the American Meteorological Society (2009) policy statement: “Geo-engineering could conceivably offer targeted and fast-acting options to reduce acute climate impacts and provide strategies of last resort if abrupt, catastrophic, or otherwise unacceptable climate change impacts become unavoidable by other means.”  The Royal Society (2009) echoed the “option of last resort” language, and invoked the value of having on hand a well-researched geo-engineering tool-kit just in case we find ourselves facing climate “tipping points.”

It is hard to argue against a safety measure, so the “just-in-case” framing reduces aversion to further investigating climate intervention technologies, and can be seen as lessening the chance of geo-engineering becoming a substitute for greenhouse gas (GHG) emission reductions.

In a sense “emergency” actions are a logical response to low probability/high consequence events. Routine risks are dealt with in routine ways (e.g., physical dividers between opposite lanes of automobile traffic) while lower probability, higher consequence, risks call for more drastic responses (e.g., on-board air bags). The highway divider operates every time a car passes to reduce the possibility that it might drift into the on-coming traffic, but the air bag is invoked only in a crash, when routine measures apparently have failed. It is a drastic response: deployment might itself harm the driver. The dividing line between routine and emergency safety measures is, of course, a fuzzy boundary, but can be found in many social responses to threat. The city of Galveston, Texas, has a sea wall to protect it from hurricane storm surges, but it also has an elaborate evacuation plan, just in case.

An emergency frame naturally raises the analog between climate change and more routine natural disasters; this gives us a place to look for lessons that might transfer to geo-engineering, a novel technology desperately in need of illumination. Two questions come quickly to mind. First, can we develop a policy system in which geo-engineering could actually work as a last ditch response? And second, does this framing reduce the likelihood that attention to geo-engineering could drain momentum from traditional global warming mitigation?

The first question falls squarely in the realm of our experience with early warning and response systems. The Royal Society (2009) report suggests that surface and near-surface albedo enhancements (via land treatments or oceanic cloud seeding) as well as stratospheric aerosol injection, could be deployed and operated on an emergency time-frame; that is, we could wait until the last minute to deploy them (on the order of a few years). The great utility is simply that this puts off deployment, allowing for more research, giving time for traditional mitigation to make a difference, and letting the need ripen to the point of obviousness. A looming climate tipping point that poses grievous harm would certainly allay qualms about the moral hazard and unintended consequences of geo-engineering. But would it evoke agreement to deploy? The Royal Society (2009) recognizes that global agreement could be hard to come by; but their deployment estimates are purely technical. Yes, we might arrange operationally to insert an aerosol cloud into the stratosphere with just a couple years of preparation, but could we get global approval of the plan in a couple of years?

Every emergency manager who has faced a hurricane knows the problem. The threat must be obvious enough, and the forecast reliable enough, to call for extra-ordinary action, like asking a million people to leave their homes. But the time required to effect the response may be large compared to the time over which the forecast is sufficiently reliable to make the decision with a reasonably low chance of over-reacting. This is the lead time problem in natural hazards, what Lenton et al. (2008), in reference to climate tipping points, called the policy time horizon.  If we are lucky, tipping points (climate emergencies) will announce themselves with enough forewarning to allow us to marshal the political and technical pieces of a geo-engineering response. The climate system may also be structured so that the largest changes exhibit relatively slow onset, allowing time for intervention to take hold. But we may be unlucky: some tipping points may offer little or no hint of their approach, may emerge quite suddenly, and may produce irreversible changes.  The lesson from disaster response is that the decision-making needs to be worked out carefully in advance; geo-engineering itself may be put off to the last moment, but that would be too late to build a workable geo-engineering governance structure.

If climate science starts to suggest untenable trade-offs between lead time, accuracy, and decision-making, it may be that the most effective path for geo-intervention would not be as last resort, but rather gradual implementation, starting sooner rather than later, based on climate trends that are not yet disastrous.  In this approach the first few increments can serve mostly for testing geo-engineering approaches, so that we better understand their effects in case more is needed.

The second, appealing, aspect of an emergency formulation of geo-engineering is that a technological fix meant to be used only in a climate emergency would seem inoculated against the possibility of infecting the global effort to reduce GHG emissions, thus avoiding the so-called moral hazard problem. Here, too, the hazards analog speaks to us: some hazards researchers have argued that not only do physical barriers like levees and sea-walls encourage more development in hazard zones (the “levee effect”), but even such emergency measures as warning and evacuation systems come to be relied upon as “routine” and, in turn, encourage risk-taking behavior. By this token the large, and remarkably effective, project of hurricane forecasting, warning, and evacuation might be tagged with worsening hurricane losses.

I should say here that I am not convinced that such a levee effect has been proven to operate in the hazard mitigation universe: the research is thin and the effect would have to hurdle a high bar: the greater losses that eventually occur would have to be discounted against the accrued  benefits of using the hazard zone. Nevertheless, could a geo-engineering response meant to be used only as last resort still produce a global levee effect, inadvertently encouraging less stringent GHG reductions? I don’t believe there’s solid evidence yet to predict such an outcome, but we should examine the possibility in any geo-engineering technology assessment. It might be that the GHG concentration increment “allowed” by the prospect of a geo-engineering fix would still pay off in net welfare. But that GHG increment might also turn out to push the climate system past a threshold into catastrophic impacts not ameliorable by last-ditch geo-engineering. That would be an emergency.

William Travis


American Meteorological Society (2009). Geoengineering the climate system: A policy statement of the American Meteorological Society. Bulletin of the American  Meteorological Society, 90:1369–1370.

Lenton, T.M.,  Hermann, H., Kreigler, E., Hall, J., Lucht, W., Rahmtorf, S. and Schellnhuber, H.J. (2008). Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Science 105: 1786-1793.

The Royal Society (2009). Geo-Engineering the Climate: Science, Governance, and Uncertainty. London.