The Antarctic Ice Sheet during the Last Glacial Maximum and its subsequent retreat history: a review
Introduction
Configuration of the Antarctic Ice Sheet during the Last Glacial Maximum (LGM), which in this paper is defined as Marine Oxygen Isotope Stage 2, and its contribution to the post LGM sea-level rise have been modeled by several investigators, with widely varying results (Fig. 1).
Stuiver et al. (1981; CLIMAP reconstruction) developed a LGM ice-sheet reconstruction that places the ice sheet at the continental shelf edge around Antarctica. Denton et al. (1991) revised the model to show little interior surface elevation change during the LGM, but considerable thickening of peripheral ice (Fig. 1). Huybrechts (1990) constructed a three-dimensional, thermo-mechanical ice-sheet model with a glacial maximum configuration smaller than that of Denton et al. (1991) and a sea-level contribution of only 12 m (Fig. 1). Glacio-hydro-isostatic models, including the ICE-3G model of Tushingham and Peltier (1991), ICE-4G model of Peltier (1994), and the ANT3 and ANT4 models of Nakada and Lambeck (1988) are based on CLIMAP reconstructions, which assume a steady-state ice sheet at the LGM (Fig. 1). These models call for approximately 20–30 m of sea-level contribution from Antarctica. The principle difference between the models is where ice occurs on the continent (Fig. 1). The results from recent models (ANT5 and ANT6) by Nakada et al. (2000) differ from previous models in that more ice exists over the Weddell Sea and the Antarctic Peninsula region (Fig. 1).
Testing the LGM models requires information about the elevation and extent of the ice sheet, and the timing of advance and retreat. Elevation information is derived from land-based studies of glacial features at elevations above the present ice-sheet surface and indirectly from raised beaches. Ice-sheet profiles are constructed from these combined data. The maximum extent of the ice sheet is recorded by glacial unconformities, geomorphic features, and sedimentary deposits on the continental shelf. Establishing the timing of ice-sheet advance and retreat is the most difficult challenge and, as a result, this remains one of the largest uncertainties.
The results from land-based studies have been summarized by Denton et al. (1991), Colhoun, 1991, Bentley and Anderson (1998), Ingolfsson et al. (1998), Miura et al. (1998), Anderson (1999), Bentley (1999). This paper focuses on marine-geological evidence for extent of the ice sheet during the LGM and on the timing of ice-sheet retreat from the shelf. The discussion concentrates on the three components of the Antarctic cryosphere, the West Antarctic Ice Sheet (WAIS), the East Antarctic Ice Sheet (EAIS), and the smaller glacial systems of the Antarctic Peninsula region (Fig. 2). A brief review of the criteria used to map the extent of the ice sheet on the shelf is provided first.
Uncertainties in radiocarbon ages, particularly with respect to the carbon reservoir effect, pose a problem with studies of this type (e.g., Andrews et al., 1999). Because of 14C recycling, the biggest source of error is with total organic carbon dates. This is complicated by the fact that the type of glacial-marine sediment from which total organic carbon dates are obtained is diatomaceous mud and ooze. These materials reveal the onset of open marine conditions, not retreat of the grounding line. The best glacial-marine sediments for dating ice sheet retreat from the shelf are those that are deposited near the grounding line, and these deposits are notoriously lacking in calcareous material. Such dates are harder to obtain, but they are worth the effort. There is general agreement that the correction factor used for calcareous marine fossils is between 1300 and 1400 yr (Gordon and Harkness, 1992; Berkman and Forman, 1996).
We have used a reservoir correction of 1300 yr for our dates, as suggested in Berkman and Forman (1996); in Table 1 we also present uncorrected dates. Dates previously published by our research group are also corrected with a 1300 correction factor. Hall and Denton (2000) compared dates from algae and co-precipitated calcium carbonate from lake sediments in the Dry Valleys and found that the dates were in agreement. We use their study as an indicator that our algal and carbonate dates should be comparable. The same 1300 yr correction (Berkman and Forman, 1996) was applied to the three dates we obtained on algal samples.
With regard to the published literature, we present radiocarbon dates as they were corrected for reservoir effects in the original source. If no correction was made in the original source, we report only the uncorrected date. Each date used in the text appears with a note on the correction used. We also reference dates obtained by surface exposure dating methods. We present these dates as they were presented by the original authors.
The maximum age that can be determined using AMS radiocarbon methods is about 46,000–50,000 yr BP (Linick et al., 1986) (http://www.physics.arizona.edu/ams/index.html). None of the dates we present exceed 40,000 yr and we have not excluded any samples for which radiocarbon dates were obtained. Additionally, for each of the dates older than 30,000 14C yr at least two dates from the same unit were obtained.
Section snippets
Criteria for mapping the LGM configuration of the ice sheet
The first seismic profiles from the continental shelf showed widespread unconformities in the younger stratigraphic section that were interpreted as glacial unconformities (Houtz and Meijer, 1970; Hayes and Davey, 1975). Mapping of these unconformities revealed a landward slope and relief that mimics the present trough and bank topography of the shelf (e.g., Alonso et al., 1992). The youngest unconformity occurs at or near the sea floor and typically amalgamates with older unconformities in a
West Antarctic Ice Sheet
The WAIS is considered more unstable than the EAIS (Hughes, 1973). It is a marine ice sheet, grounded well below sea level and characterized by rapid flow and discharge relative to the EAIS. Most of this discharge occurs through ice streams that reach flow velocities of hundreds of meters per year (Fig. 2). Marine-geological research conducted in recent years has focused on the principle drainage outlets of the WAIS (Shipp et al., 1999; Anderson and Shipp, 2001; Wellner et al., in press).
Results
Our LGM reconstruction is shown in Fig. 13. The grounding line location shown is a minimum grounding line that is based on the presence of subglacial geomorphic features and/or till. A maximum grounding line at the shelf margin is inferred from a glacial unconformity that extends to the shelf break in many areas. However, the age of this unconformity is uncertain.
The marine data place constraints on the expansion (and therefore coastal thickness) of the ice sheet during the LGM. Our LGM
Conclusions
- 1.
During the LGM, the WAIS advanced across the continental shelf, and reached the shelf margin in the central and eastern Ross Sea. Ice advanced at least to the middle continental shelf along the rest of the WAIS margin.
- 2.
Initial retreat of the WAIS appears to have occurred between 15,000 and 12,000 yr and continues into the Holocene. There was significant retreat of the ice sheet in the Ross Sea, and possibly Weddell Sea, after 7000 yr BP.
- 3.
During at least the final stages of the glacial maximum, the
Acknowledgements
This research was funded by the National Science Foundation, Office of Polar Programs grant number OPP-9527876. Radiocarbon dating was performed at the University of Arizona AMS laboratory. We appreciate the input from two anonymous reviews and editor Peter Clark.
References (90)
- et al.
Sedimentation on the Ross Sea continental shelf, Antarctica
Marine Geology
(1984) - et al.
Problems and possible solutions concerning radiocarbon dating of surface marine sediments, Ross Sea, Antarctica
Quaternary Research
(1999) - et al.
Holocene raised beaches at Terra Nova Bay, Victoria Land, Antarctica
Quaternary Research
(1991) Volume of the Antarctic ice at the last glacial maximum, and its impact on global sea level change
Quaternary Science Reviews
(1999)- et al.
Late Quaternary glacial history of George VI Sound area, West Antarctica
Quaternary Research
(1982) - et al.
Application of tandem accelerator mass-spectrometer dating to late Pleistocene–Holocene Sediments of the East Antarctic continental shelf
Quaternary Research
(1989) - et al.
Seismic stratigraphic evidence of ice-sheet advances on the Wilkes Land margin of Antarctica
Sedimentary Geology
(1995) - et al.
Glacio-isostasy and glacial ice load at Law Dome, Wilkes Land, East Antarctica
Quaternary Research
(2000) - et al.
Developing sediment geochronologies for high-latitude continental shelf deposits: a radiochemical approach
Marine Geology
(1992) - et al.
Bottom currents, sedimentation and ice-sheet retreat facies successions on the Mac Robertson shelf, East Antarctica
Marine Geology
(1998)