Long integrations of the Antarctic ice sheet during the last glacial cycles were analysed in Budd et al. (1998), Huybrechts & de Wolde
(1999), Ritz et al. (2001) and Huybrechts (2002). Plate 80.3 (left panel) shows the evolution of key glaciological variables over the last four glacial cycles in a typical run with the Huybrechts model, with forcing derived from the Vostok ice core (Petit et al., 1999) and the SPECMAP sea-level stack (Imbrie et al., 1984). In line with the generally accepted view, volume changes are largely concentrated in the West Antarctic and Peninsula ice sheets. These are caused by a repeated succession of areal expansion and contraction of grounded ice close to the continental break during glacial maxima. Around the East Antarctic perimeter, grounding-line advance was limited because of the proximity of the present-day grounding line to the continental shelf edge. In these models, glacial-interglacial fluctuations are mainly controlled by changes in the global sea-level stand and dynamic processes in the ice shelves. This supports the hypothesis that the Antarctic ice sheet basically follows glacial events in the Northern Hemisphere by means of sea-level teleconnections. Typical glacial-interglacial volume changes correspond to global sea-level contributions of about 20 m. Freshwater fluxes originating from the Antarctic ice sheet are an important output because of their role in modulating the deep-water circulation of the ocean. Model predictions displayed in Plate 80.2 show that these are fairly constant in time and are almost entirely dominated by the iceberg flux. During the last two glacial-interglacial transitions meltwater peaks occurred about three times larger than the normal background fluxes. During interglacials, melting from below the ice shelves is also an important contribution but surface runoff always remained negligible.
According to the model, surface elevations over most of West Antarctica and the Antarctic Peninsula were, at the Last Glacial Maximum (LGM), up to 2000m higher than at present in direct response to the grounding-line advance (Plate 80.4). Over central East Antarctica, surface elevations at the LGM were 100-200 m lower because of the lower accumulation rates (Huybrechts, 2002). A characteristic of this model is that most of the Holocene grounding-line retreat in West Antarctica occurs after 10kyrBP and lags the eustatic forcing by up to 10kyr. This behaviour is related to the existence of thresholds for grounding-line retreat, and to the offsetting effect of the late-glacial warming leading to enhanced accumulation rates and a thickening at the margin. The late timing is in line with recent geological evidence (Ingolffson et al., 1998; Conway et al., 1999) and is supported by some interpretations of relative sea-level data (Tushingham & Peltier, 1991), but other inferences have been made. The implication is an ongoing shrinking of the Antarctic ice sheet at the present time equivalent to a global sea-level rise of about 2.5cm per century (Huybrechts & de Wolde, 1999). An important unknown regarding the glacial history of the West Antarctic ice sheet is whether widespread ice-streaming comparable to the present Siple Coast continued to exist at LGM, in which case surface elevations may have been substantially lower than shown in Plate 80.4, and the contribution to the global sea-level lowering was less by perhaps several metres (Huybrechts, 2002).
Was this article helpful?