The geometry of the Antarctic Ice Sheet throughout the last glacial cycle has been compiled using a number of different methods. In terrestrial environments, the aerial extent of former ice cover is generally recognised by mapping the lateral extent of ice marginal landforms such as moraines or ice marginal lake sediments, or by mapping subglacial landforms such as drumlins and striae. Ice volume can equally be reconstructed by mapping glacial debris striae, the locations of erratic boulders and the height/extent of mountain trimlines. The relative age of each mapped unit is usually constrained using a measure of exposure age, most often by the degree of weathering (such as soil formation or development of tafoni or iron staining on boulders), and more recently by measurement of isotopes produced by cosmogenic radiation. Stratigraphic correlations are generally difficult due to lack of natural exposures.
Similar techniques are also used in marine environments, but rather than investigating the geomorphology of former glacial landscapes using field mapping and aerial photography, the ocean floor is imaged using techniques such as swath bathymetry and side-scan sonar. These techniques commonly identify ice marginal landforms such as moraines, or subglacial landforms such as mega scale glacial lineations, meltwater channels and drumlins. Stratigraphic techniques are more useful in the marine environment due to the ease of data acquisition via acoustic and seismic surveys, particularly when these sediment units are also sampled by gravity or piston cores.
Establishing detailed late Quaternary glacial chronologies in Antarctica can be challenging (e.g., Anderson et al., 2002; Ingolfsson, 2004). The most direct dating is usually achieved in presently terrestrial environments, through cosmogenic exposure dating of moraines or radiocarbon dating of fossils contained within emergent marine sediments or ice-marginal lakes. However, factors such as production rate uncertainties (e.g., Gosse and Phillips, 2001), recycling of clasts with prior exposure and post-depositional reworking of glacial sediments (e.g., Brook et al., 1995) reduce the precision of cosmogenic chronologies to around 10% (Putkonen and Swanson, 2003). Where such materials are preserved, the algal mats commonly found in proglacial lake sediments can provide reasonably accurate radiocarbon ages, provided that the water column is well mixed and that at least part of the lake surface ice cover melts out each summer (Gore 1997b; Hendy and Hall, 2006). Limiting ages on glacial events in terrestrial environments can sometimes also be constrained by radiocarbon dating of the small amount of organic material sometimes found in ice-marginal or postglacial sediments (e.g., Burgess et al., 1994).
Carbon sourced from marine environments, such as the bodies of seals and penguins or carbonate shells in former (and emergent) marine environments is subject to a marine reservoir effect that varies around the continent, but is on the order of 1.3 ka (e.g., Berkman and Forman, 1996; but also see Kiernan et al., 2003).
On the continental shelf the majority of published data with regard to the timing of ice advance and retreat have been obtained by Atomic Mass Spectrometry (AMS) radiocarbon dating of the acid-insoluble (mostly diatom) fraction of organic material that is preserved immediately above or below a layer of glacially derived sediment. There are several potential problems with this technique, for example the influx of ''dead'' carbon from ice melt, fine-grained geogenic carbon such as graphite and recycling of older organic matter means that such material at the modern sediment/water interface provides radiocarbon ages of >2 ka, and sometimes > 6 ka in areas of little biological productivity (e.g., Andrews et al., 1999). These values exceed the age derived from carbonate shells from living biota at such sites by over 1.3 ka. To some extent this problem has been rectified in more recent studies by AMS analysis of discrete carbonate shells (e.g., Leventer et al., 2006), but the lack of shells in many cores means that this approach is not always possible. Despite these difficulties, considerable efforts have been made in order to constrain the timing of ice advance and retreat across the continental shelf in many sectors of the continent.
Was this article helpful?