Evidence of changes in icesheet geometry and abrupt changes in mass balance

If the LIS had been erosive at all points under its bed then the amount of information on past flow directions would be nonexistent. Far travelled glacial erratics (see above) would provide some information but these are not imprinted on the bed of the ice sheet, hence relative changes in flow regimes could not be ascertained. With the advent of LANDSAT and other highresolution imagery several research groups have developed procedures for mapping the relative ages of different flow directions on the bed of the former ice sheet (Boulton & Clark, 1990; Clark, 1993, 1997; Kleman & Hattestrand, 1999). These papers suggest a more dynamic LIS than might be judged from the relatively uncomplicated pattern of ice retreat since the LGM, which shows a retreat of the LIS towards the north and east, thus towards the ancestral homeland on the uplands of the eastern Canadian Arctic (Andrews, 1973). Indeed, the Barnes Ice Cap is a relic of the LIS. The mapping of ice-flow directions from imagery is also being combined with fieldwork to verify the interpretations, however some of the interpretations, especially in the area of Labrador-Ungava, are controversial (Veillette et al., 1999).

Up until the very early 1990s researchers often talked about the possible past and future collapse of the West Antarctic Ice Sheet (WAIS) (Mercer, 1978; Thomas et al., 1979; Weertman & Birchfield, 1982; MacAyeal, 1992a). However, in 1988 (Heinrich, 1988) and subsequently (Andrews & Tedesco, 1992; Andrews, 1998; Bond et al., 1992; Broecker et al., 1992; Hesse & Khodabakhsh, 1998) it was shown that massive collapses of the LIS had occurred during the Wisconsinan glaciation in what are now referred to as Heinrich (H-) events. These have been labelled and dated at ca. 13 (H-0), 16.5 (H-1), 24 (H-2), 29 (H-3) and 40 (H-4) ka, with two older events, H-5 and H-6, at 46 and 60ka (Chapman & Shackleton, 1999; Bond et al., 1999). These events were centred on Hudson Strait (MacAyeal, 1993a), and largely involved the transport of glacially eroded detrital carbonate from the floors of Hudson Strait, and possibly Hudson Bay, to sites as far distant as the deep-sea basins off Ireland and even south to near Portugal (Lebreiro et al., 1996). The cause(s) of H-events is not known with certainty but may have involved a 'binge-purge' cycle associated with changes in ice-stream thickness, activity, and basal temperatures (MacAyeal, 1993a, Alley & MacAyeal, 1994; Clarke et al., 1999). Although the literature most often stresses the iceberg rafted (IRD) component, defined as lithics >125 |lm, it is critical to note that in the ice-proximal area below the shelf-break off Hudson Strait the sediments are fine-grained, often laminated and produced by turbidites (Hesse et al., 1997, 2004). The evidence suggests that massive outburst floods were part and parcel of the conditions associated with H-events.

A variety of tracers are now being used to better understand where the sediments associated with H-events came from on the bed of the LIS. The tracers include 40/39Ar dates on individual hornblende grains (Hemming et al., 2000, 2002a), the isotopic composition of the <63 |m decalcified sediment fraction (Farmer et al., 2003), the petrology of the sand fraction (Bond & Lotti, 1995), the rare earth composition of the sediments (Benson et al., 2003), and rock magnetic properties (Stoner et al., 1996; Stoner & Andrews, 1999). The recent work of Farmer et al. (2003) from the eastern margin of the LIS was able to characterize the glacially transported sediments in a series of offshore marine cores and link them with source areas. An important aspect of this work was the demonstration that the bulk of the sediment was derived from the adjacent continent. Down-core studies (Hemming et al., 2002a; Groussett et al., 2001) indicate that these various approaches will capture changes in the flow directions of the ice sheet and hence offer the possibility of linking the geomorpho-logical bed record, which is on a relative dating scale, with a more numerical age scale based on the rate of sediment accumulation in marine cores. An exciting prospect!

0 0

Post a comment