IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis Melting and Calving of Ice Shelves

Many of the largest and fastest ice sheet changes thus appear to be at least in part responses to ice shelf shrinkage or loss. Although ice shelf shrinkage does not directly contribute to sea level change because shelf ice is already floating, the very tight coupling to inland ice means that ice shelf balance does matter to sea level. The available data suggest that the ice shelf changes have resulted from environmental warming, with both oceanic and atmospheric temperatures important, although changes in oceanic circulation cannot be ruled out as important contributors.

The southward-progressing loss of ice shelves along the Antarctic Peninsula is consistent with a thermal limit to ice shelf viability (Morris and Vaughan, 2003). Cook et al. (2005) found that no ice shelves exist on the warmer side of the –5°C mean annual isotherm, whereas no ice shelves on the colder side of the –9°C isotherm have broken up. Before the 2002 breakup of the Larsen B Ice Shelf, local air temperatures had increased by more than 1.5°C over the previous 50 years (Vaughan et al., 2003), increasing summer melting and formation of large melt ponds on the ice shelf. These likely contributed to breakup by draining into and wedging open surface crevasses that linked to bottom crevasses filled with seawater (Scambos et al., 2000). Large ice flow models do not accurately capture the physical processes involved in such dramatic iceberg calving, or in more common calving behaviour.

Despite an increased ice supply from tributary glaciers, thinning of up to several metres per year has been measured for ice shelves on the Amundsen Sea coastline in the absence of large surface mass balance changes. This suggests that increased basal ice melting is responsible for the thinning (Shepherd et al., 2003, 2004). Similarly, the 15-km floating ice tongue of Jakobshavn Glacier survived air temperatures during the 1950s similar to or even warmer than those associated with thinning and collapse near the end of the 20th century, implicating oceanic heat transport in the more recent changes, although air temperature increases may have contributed (Thomas et al., 2003).

The basal mass balance of an ice shelf depends on temperature and ocean circulation beneath it. Isolation from direct wind forcing means that the main drivers of circulation below an ice shelf are tidal and density (thermohaline) forces. Lack of knowledge of sub-ice bathymetry has hampered the use of three-dimensional models to simulate circulation beneath the thinning ice shelves. Both the west side of the Antarctic Peninsula and the Amundsen Sea coast are exposed to warm Circumpolar Deep Water (CDW; Hellmer et al., 1998), capable of causing rapid ice shelf basal melting. Increased melting in the Amundsen Sea is consistent with observed recent warming by 0.2°C of ocean waters seaward of the continental shelf break (Jacobs et al., 2002; Robertson et al., 2002). Simple regression analysis of available data including those from the Amundsen Sea indicated that 1°C warming of waters below an ice shelf increases basal melt rate by about 10 m yr–1 (Shepherd et al., 2004).