19.3.5.2 Deglaciation of West Antarctic and Greenland ice sheets

The potential for partial or near-total deglaciation of the Greenland and the West Antarctic ice sheets (WAIS) and associated sea-level rise (Jansen et al., 2007 Sections 6.4.3.2 and 6.4.3.3; Meehl et al., 2007 Sections 10.6.4, 10.7.4.3 and 10.7.4.4; Alley et al., 2005; Vaughan, 2007), is a key impact that creates a key vulnerability due to its magnitude and irreversibility, in combination with limited adaptive capacity and, if substantial deglaciation occurred, high levels of confidence in associated impacts. Ice sheets have been discussed specifically in the context of Article 2 (O’Neill and Oppenheimer 2002; Hansen, 2005; Keller et al., 2005; Oppenheimer and Alley, 2005). Near-total deglaciation would eventually lead to a sea-level rise of around 7 m and 5 m (***) from Greenland and the WAIS, respectively, with wide-ranging consequences including a reconfiguration of coastlines worldwide and inundation of low-lying areas, particularly river deltas (Schneider and Chen, 1980; Revelle, 1983; Tol et al., 2006; Vaughan, 2007). Widespread deglaciation would not be reversible except on very long time-scales, if at all (Meehl et al., 2007 Sections 10.7.4.3 and 10.7.4.4). The Amundsen Sea sector of the WAIS, already experiencing ice acceleration and rapid ground-line retreat (Lemke et al., 2007 Section 4.6.2.2), on its own includes ice equivalent to about 1.5 m sea-level rise (Meehl et al., 2007 Section 10.7.4.4; Vaughan, 2007). The ability to adapt would depend crucially on the rate of deglaciation (**). Estimates of this rate and the corresponding time-scale for either ice sheet range from more rapid (several centuries for several metres of sea-level rise, up to 1 m/century) to slower (i.e., a few millennia; Meehl et al., 2007 Section 10.7.4.4; Vaughan and Spouge, 2002), so that deglaciation is very likely to be completed long after it is first triggered.

For Greenland, the threshold for near-total deglaciation is estimated at 3.2-6.2°C local warming (1.9-4.6°C global warming) relative to pre-industrial temperatures using current models (Meehl et al., 2007 Section 10.7.4.3). Such models also indicate that warming would initially cause the Antarctic ice sheet as a whole to gain mass owing to an increased accumulation of snowfall (*; some recent studies find no significant continent-wide trends in accumulation over the past several decades; Lemke et al., 2007 Section 4.6.3.1). Scenarios of deglaciation (Meehl et al., 2007 Section 10.7.4.4) assume that any such increase would be outweighed by accelerated discharge of ice following weakening or collapse of an ice shelf due to melting at its surface or its base (*). Mean summer temperatures over the major West Antarctic ice shelves are about as likely as not to pass the melting point if global warming exceeds 5°C (Meehl et al., 2007 Section 10.7.4.4). Some studies suggest that disintegration of ice shelves would occur at lower temperatures due to basal or episodic surface melting (Meehl et al., 2007 Sections 10.6.4.2 and 10.7.4.4; Wild et al., 2003). Recent observations of unpredicted, local acceleration and consequent loss of mass from both ice sheets (Alley et al., 2005) underscores the inadequacy of existing ice-sheet models, leaving no generally agreed basis for projection, particularly for WAIS (Lemke et al., 2007 Section 4.6.3.3; Meehl et al., 2007 Sections 10.6.4.2 and 10.7.4.4; Vieli and Payne, 2005). However, palaeoclimatic evidence (Denman et al., 2007 Sections 6.4.3.2 and 6.4.3.3; Overpeck et al., 2006; Otto-Bliesner et al., 2006) suggests that Greenland and possibly the WAIS contributed to a sea-level rise of 4-6 m during the last interglacial, when polar temperatures were 3-5°C warmer, and the global mean was not notably warmer, than at present (Meehl et al., 2007 Sections 10.7.4.3 and 10.7.4.4). Accordingly, there is medium confidence that at least partial deglaciation of the Greenland ice sheet, and possibly the WAIS, would occur over a period of time ranging from centuries to millennia for a global average temperature increase of 1-4°C (relative to 1990-2000), causing a contribution to sea-level rise of 4-6 m or more (Meehl et al., 2007 Sections 10.7.4.3 and 10.7.4.4; Oppenheimer and Alley, 2004, 2005; Hansen, 2005).

Current limitations of ice-sheet modelling also increase uncertainty in the projections of 21st-century sea-level rise (Meehl et al., 2007 Section 10.6.4.2) used to assess coastal impacts in this report. An illustrative estimate by WGI of the contribution of processes not represented by models yielded an increase of 0.1-0.2 m in the upper ranges of projected sea-level rise for 2100 (Meehl et al., 2007 Section 10.6.4.2). Other approximation methods would yield larger or smaller adjustments, including zero.