Antarctic Sea Ice Drilling

Progress in further sea-ice drilling began slowly with a SCAR workshop in Bremerhaven in 1990 for discussions that led to the formation of SCAR Group of Specialists on Global Change (GLOCHANT). While Cenozoic glacial history was seen as beyond the remit of the group, the meeting was productive in that it began a discussion among US, NZ, Italian, German and UK scientists for a workshop in Wellington (Barrett and Davey, 1992) that initiated the Cape Roberts Project (CRP). Following the meeting, national programmes agreed to plan for drilling four holes in two seasons to core a 1,500 m sequence that was thought from seismic correlation to extend the CIROS-1 record back to the Cretaceous. A new drilling system was also needed (Fig. 3.6).

Figure 3.6: Diagram showing the advances made in sea riser design after the first decade of McMurdo Sound drilling for the Cape Roberts Project, and the further developments that were required for ice-shelf drilling (Modified from Harwood et al., 2003, with permission from the ANDRILL Science Management Office).

Figure 3.6: Diagram showing the advances made in sea riser design after the first decade of McMurdo Sound drilling for the Cape Roberts Project, and the further developments that were required for ice-shelf drilling (Modified from Harwood et al., 2003, with permission from the ANDRILL Science Management Office).

The Cape Roberts Project sites were drilled, after a year's delay on account of poor ice conditions, and with three drilling seasons (1997-1998 to 1999-2000, reported in CRST (Cape Roberts Science Team) (1998, 1999, 2000) and Davey et al. (2001). The 1,500 m-thick section cored provided the first proximal high-resolution record from the Antarctic continental shelf of climate history for the period from 34 to 17 Ma. A combination of biomagnetostratigraphic dating with Ar-Ar radiometric ages at several key points provided robust (<0.5 million years) resolution for most of the interval and excellent resolution (<0.1 million years) for three cycles around 24 Ma ago (latest Oligocene) (Florindo et al., 2005, but see update by Naish et al., 2008b). These showed for the first time that the Antarctic Ice Sheet was responding to orbital forcing with 40,000 and 100,000 year frequencies in the distant past (Naish et al., 2001). The cores also recorded the influence of orbital forcing on Antarctic Ice Sheets through Oligocene and early Miocene times as well as persistent slight cooling over this period (Barrett, 2007; Dunbar et al., 2008). This contrasted with the late Oligocene warming deduced from a review of deep-sea oxygen isotope data by Zachos et al. (2001), but was consistent with the reinterpretation of oxygen isotope data sets for this period by Pekar et al. (2006).

3.6.2. Ice Sheet and Climate Modelling

ANTOSTRAT in a few short years had greatly extended the significance and value of the few drill holes that could be drilled on the Antarctic margin by providing a basis for correlating events from basin to basin. However, the behaviour of the ice that eroded and deposited the sediments was not well understood, especially on a continental scale. It was assumed that on a roughly circular continent, the extent of the ice would be similar in all directions, but Antarctica has enough irregularities, mountains and basins to see that local geological histories could vary. Ice-sheet modelling provided a means of objectively experimenting on and visualizing ice-sheet behaviour, Oerlemans (1982) providing an early example.

Subsequent ice-sheet modelling by Huybrechts (1993) (Fig. 3.7) was significant because it showed a clear and consistent relationship between temperature and ice volume, including a slight increase in ice mass with initial warming, thus linking past Antarctic temperatures and sea level change. It indicated the loss of West Antarctic ice with a regional warming of 10°C, in the range Mercer (1978) had suggested, and provided patterns of ice growth and decay that varied around the continent with changes in temperature. The model also supported the view that the earliest ice sheets

Figure 3.7: Modelling the Antarctic Ice Sheet for progressively higher regional temperatures (Huybrechts, 1993). If a polar amplification factor of 2 is assumed (Manabe and Stouffer, 1980), then the effective rise in average global temperature to achieve each of these states would be 1/2 of the temperature rise shown.

Figure 3.7: Modelling the Antarctic Ice Sheet for progressively higher regional temperatures (Huybrechts, 1993). If a polar amplification factor of 2 is assumed (Manabe and Stouffer, 1980), then the effective rise in average global temperature to achieve each of these states would be 1/2 of the temperature rise shown.

would have reached the coast in the Prydz Bay, whereas the Transantarctic Mountains offered a significant barrier ice reaching to the Ross Sea.

3.6.3. Discoveries in the Transantarctic Mountains Reinterpreted

Huybrechts' model also suggested that it was not possible to create conditions in which an ice sheet could grow and over-top the present Transantarctic Mountains at their present height, depositing wet-based glacial debris at the margins. Some argued that the mountains were lower in the Pliocene and had risen rapidly since then (Behrendt and Cooper, 1991), but this was subsequently countered in a review of Antarctic tectonic history by Fitzgerald (2002). However, there were other substantive concerns. New high-resolution deep-sea isotope records of the last few million years gave no indication of negative anomalies to be expected from extensive ice-sheet recession implied by Pliocene marine basins in the East Antarctic interior (Denton et al., 1991; Kennett and Hodell, 1993). Marchant et al. (1993) reported volcanic ashes at high elevations in the McMurdo Dry Valleys and dating back to ~ 11 million years, whose survival seemed inconsistent with a warm Pliocene and an ice-free Antarctic interior. Also, diatoms were discovered in South Pole snow (Kellogg and Kellogg, 1996) and surficial rock debris, and counts showed they were extremely rare in Sirius deposits (Barrett et al., 1997). These observations led to the view that the Sirius deposits predated the widely accepted mid-Miocene cooling, the few age-critical Pliocene diatoms resulting from atmospheric transport. The report by Gersonde et al. (1997) of a Pliocene meteorite impact event ejecting diatomaceous sediment into the stratosphere from the southeastern Pacific Ocean floor provided another source for possible atmospheric contamination. However, Harwood and Webb (1998) maintained that the marine diatoms and diatomite clasts found by Harwood were larger than winds could carry.

In reviewing Antarctic climate through Cenozoic times, Barrett (1996. 1999) noted that McMurdo Sound drill cores had shown Oligocene and early Miocene Antarctic Ice Sheets were of similar extent to those of recent times during glacial periods, with largely ice-free forested coasts during interglacial periods, thus resembling the dynamic behaviour of Northern Hemisphere ice sheets in the Quaternary. This contrasted with the relative stability of the Antarctic Ice Sheet at this time, when there has been little change in its central elevation, although a significant volume of ice has been lost around the margin, ~ 15 m of sea level equivalent (SLE) since the Late Glacial Maximum (Zwartz et al., 1997). While offshore mid-Miocene to early Pliocene strata had yet to be cored, seismic stratigraphy of the continental shelf showed numerous sediment packages and unconformities indicating significant glacial deposition and erosion extending to the shelf edge. These data implied significant dynamism of the ice sheet at the margins (Anderson and Bartek, 1992), but indicated little about its interior health. The evidence cited above for a persistent late Neogene ice sheet in the East Antarctic interior was not necessarily inconsistent with clear evidence of Pliocene warmth from marginal coastal marine sediments, as in the region east of Prydz Bay (Harwood et al., 2000).

0 0

Post a comment