THE LONG TAIL ON SEA-LEVEL RISE: DEEP UNCERTAINTY ABOUT LARGE RISKS

Sea-level rise is virtually certain with warming, and rise could be several times larger than generally projected. IPCC projections of approximately 1 m of rise by 2100 under the strongest warming considered are dominated by expansion of warming ocean water and by transfer of water to the oceans primarily from melting of mountain glaciers and the edge of the Greenland Ice Sheet. Some published studies suggest a tendency for actual rise from these sources to exceed projections, but the errors likely are relatively small and well-characterized. If well-known but poorly modeled additional physical processes are triggered especially in West Antarctica, however, the sea-level rise could be much larger. Fracture processes leading to ice-shelf loss and calving-cliff breakage likely would be important in rapid rise, and could be triggered only if certain thresholds are crossed. Most of the models cited in support of the IPCC projections have omitted representation of these processes, have not been tested against the full range of paleoclimatic data, and so cannot be used to assess the likelihood of large, rapid sea-level rise. Modeling that incorporates more of the processes and parameterizes against paleoclimatology can produce rapid rise under strong warming, but with deep uncertainty. From his student days, the author recalls a metallurgical-engineering professor urging exceedingly precise calculations followed by engineering design with a threefold factor of safety, because of the likelihood of factor-of-two errors in dealing with fractures. Policy advice under such circumstances is difficult, but both mitigation and flexible design are likely to be much more valuable than under the usual IPCC projections.