Analyzing Plausible Extreme Event Scenarios

Climate scientists have paid significant attention to some of the ways that gradual global climate change might lead to abrupt and sometimes irreversible large changes at a continental or regional scale (National Research Council, 2002; Alley et al., 2003; Lenton et al., 2008). None of these possibilities can be projected with a level of confidence that would satisfy climate scientists. However, from a security analysis standpoint, in which attention is paid to future scenarios that are of sufficiently high consequence even if their probabilities are relatively low, this scientific work points to particular possible extreme event scenarios that are worthy of further analysis.

An example of the kind of process that could lead to surprising and very extreme events can be drawn from evidence in the paleoclimate records combined with recognition of enhanced polar temperature variations due to changes in GHG concentrations. Citing an observation by Bintanja et al. (2005) that over the past 800,000 years a 1°C increase in global mean temperature was associated with increased equilibrium sea levels of about 20 meters, Hansen and Sato (2012) have suggested that the sea level rise in the next century may well be on the order of 5 meters. They argue that an increase of 3.6°F (2°C) over pre-industrial temperature levels, which is highly likely to occur in this century, would commit the planet to sea level rise of many meters. Given the considerable uncertainty in the science of glaciology about the stability of major ice sheets, it is unclear whether their contribution to sea level rise over the next century will be linear or will follow a nonlinear trajectory with an increasing rate of change over time. If nonlinear processes prevail, then the common projection of up to 1 meter by the end of the century may be a lower bound rather than an upper bound. The rate at which the sea level rise would occur is critically important, of course, in terms of the social and political consequences.

To better evaluate the import for U.S. national security of scenarios like this, which have some scientific plausibility but which extend beyond the current scientific consensus, the intelligence community might benefit from several types of knowledge that could be developed in the coming decade to help analysts anticipate security issues that might arise if such a scenario is realized. These would include improved measures of rates of change in temperature and glacier ice cover in the polar regions; the use of existing climate models to project how this degree of ice melting would affect such outcomes as coastal inundation, extreme precipitation, and cyclonic storm severity; and assessments of the exposure, vulnerability, and response capacity of key countries and regions to these outcomes.

Several other examples of potential rapid-onset extreme climate event scenarios can readily be found (Lenton et al., 2008). For instance, models of changes in the Indian summer monsoon indicate that several sharply dif-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement