It is known that the Antarctic ice sheet greatly affected global climate, sea-level, ocean circulation, and Southern Hemisphere biota during Cenozoic times, but until recently, how remained largely a mystery. Because few Cenozoic rocks are exposed on the Antarctic continent, the only way to investigate the evolution of the Antarctic ice sheet was through isotopic studies on distant deep-ocean sediments. The success of these studies was limited_isotopic studies and low-latitude sea level variation yielded different glacial ice volumes.
To take study of Cenozoic Antarctic glaciations and their effects a step further, the Antarctic Offshore Acoustic Stratigraphy project (ANTOSTRAT) assembled a contingent of geoscientists from over 100 institutions in 32 countries. Since 1989, they have compiled and analyzed offshore acoustic and geologic data from the Antarctic continental margin, amassing over 150,000 km of multichannel seismic reflection data, tens of thousands of kilometers of single-channel, high-resolution seismic data, and over 1000 cores.
In August 1994, a symposium in Siena, Italy, was held to examine results of the ANTOSTRAT studies within five broad segments of the Antarctic margin: Wilkes Land, Prydz Bay, Weddell Sea, Antarctic Peninsula, and Ross Sea. The 85 participants debated important thematic issues of Cenozoic glacial history and evolution of the vast offshore sedimentary records and developed strategies for future Antarctic drilling and coring.
The symposium began with reviews of scientific disciplines and their stands on current issues: conflicting Antarctic glacial histories (a massive stable ice sheet versus many fluctuating ice sheets); ambiguities in determining glacial sediment facies from seismic data using existing sediment cores; conflicting models of controls of ice sheet and sediment dispersal by glaciological factors and underlying topography; and lithospheric response to cyclic ice and sediment loading under different tectonic regimes.
Regional reviews and detailed studies made clear that the grounded Antarctic ice sheet, which is greatly affected by sea level and underlying topography, has strongly controlled deposition and erosion of offshore sedimentary sequences. The similar geometries of Cenozoic sequences beneath all segments of the Antarctic margin suggest that climatic factors, such as glacial history and ocean circulation, exerted greater control on sequence geometry than local tectonics. In several areas, the connection between sediment and ice stream transport was strengthened: shelf-edge deposition is concentrated where broad troughs crossing the shelf from the continental interior mark the paths of ice streams during glacial maxima.
Regional studies also refined and extended previous work. Where seismic data are extensive_for example, in the Bransfield Strait and Ross Sea_detailed mapping illustrates extensional basement structures that control ice sheet locations and directions of episodic ice stream flow and sediment dispersal. Elsewhere, large-scale features within the glacial sequences extend the range of glacial features that need to be explained. Examples of these include unconformities that cross the entire margin in many areas and giant slides down the continental slope in the Weddell Sea. Kilometer-relief terrigenous sediment drifts on the Antarctic Peninsula continental rise provide an additional depositional environment with a record of glacial history. High- resolution studies revealed details of ice sheet control of sedimentation during the Holocene deglaciation. However, the unusually cold Antarctic ice sheet in the Pleistocene, which experienced a large externally imposed sea-level signal from Northern Hemisphere glaciation, may not be an entirely appropriate model for earlier glaciations.
Fig. 1. Maps of the Antarctic ice sheet at present and glacial maximum conditions, showing ANTOSTRAT study areas, where acoustic and geologic data are being used to study interrelationships of grounded Cenozoic ice sheets and sea levels.
A forum on polar seismic stratigraphy centered on the extent to which glacially derived sediments could be unambiguously identified on seismic profiles to guide future seismic data acquisition and assess priorities for direct sampling. Discussion revolved fundamentally around scale. Detailed mapping of some Northern Hemisphere shelves was used to argue that some very small-scale features are unambiguously glacial. Recognizing these small features within seismic sequences required high-frequency survey with small line spacing, which is not effective at great depth and not always feasible in the Antarctic. Although higher resolution data have greater diagnostic power, some characteristics of glacial sedimentation were recognizable at all frequencies: notably the large-scale geometry of an overdeepened (200-1200 m deep), inward-sloping shelf.
Glacial environments have highly varied sedimentary expression. Temperate glaciers, with abundant basal melt-water and sediment supply, are characterized by acoustically well-stratified marine units. The record of polar glaciers, with negligible basal melt-water and sediment outwash, is dominated by massive subglacial facies. The extraordinarily large-volume Antarctic prograding-sediment sequences have been attributed to sediment transport within a deforming till layer at the base of fast-moving grounded ice streams, like the 50 km-wide ice stream B in the Ross Embayment. Besides depositing sediment at their termini, such ice streams may widely erode the continental shelf with greatest erosion on the inner shelf.
Participants discussed acoustic features of glaciated shelves. Polar ice streams carve 20-100-km-wide shallow troughs that differ from the narrow fluvial valleys and slope canyons of nonglacial margins. Coastal onlap, coastline and facies migration, fluvial incision, or transgressive- regressive sequences generally do not occur on the 200-1200-m-deep continental shelf, because 50-100 m sea-level fluctuations do not subaerially expose the shelf. Regionally flat erosional unconformities on the shelf far below wave base suggest erosion by extensive grounded ice sheets during times when the shelf was overdeepened. Clinoforms with 5 -17° dips and 1-km-relief suggest extraordinarily high sediment flux possibly from a basal deforming till layer. Elongate, unstratified asymmetric mounded facies are possible morainal deposits, and linear sea-floor furrows and multiple, thin, over-consolidated beds indicate iceberg grounding and likely ice-sheet loading.
Not all of these glacial features would be present in a particular area, and some might be difficult to identify at depth. In many places, evidence of uniformly stratified units indicates that deposition was in open-water marine conditions, without grounded ice sheets. Participants agreed that results from acoustic stratigraphic inferences of fluctuations of former grounded ice sheets and sea levels would be speculative until additional coring and drilling is done to validate the results.
A variety of useful drilling methods promise to increase our knowledge at different depths. An ice-platform drilling system that can core 500+ mbsf inshore with 98% recovery will be used in the southwest Ross Sea to core presumed Oligocene and older strata. Piston coring will be used with both conventional and hydraulic corers for 10-m maximum penetration on the continental shelf. "Over the (ship) side" drill rigs placed on floating ice can drill 50 mbsf with high recovery. Most participants agreed that 50 mbsf minimum penetration was needed, preferably from a ship-based platform for mobility, to assure sampling beneath ubiquitous surficial diamicton.
Regional working groups identified coring and drilling sites for a coordinated examination of ice sheet variability. For the Weddell Sea and Prydz Bay, long drilling transects were proposed close to axes of large, well-defined fans derived from major East Antarctic ice-sheets since the Eocene. Sampling at these transects can divulge information about the Paleogene glacial onset and the successive Neogene cooling intervals. A transect of 50-m cores across the deeply eroded Wilkes Land shelf promises to provide a complementary proximal record. Prograding sequences in the northwestern and eastern Ross Sea would help resolve detailed marginal histories of East and West Antarctic ice sheets and their relationship to each other. Existing Ross Sea drill sites on the middle and inner shelf provide only fragments of a general glacial history. Participants plan to submit proposals to ODP in addition to drill sites that have already been proposed this year to obtain a high-resolution, late Cenozoic record for the Antarctic Peninsula as part of a circum-Antarctic program focused on ice sheet history and global sea level change.
Papers, atlas maps, CD-ROMs, and other scientific results of ANTOSTRAT and the symposium are planned for publication in two forthcoming volumes of the AGU Antarctic Research Series. The symposium was convened by Alan Cooper, Peter Webb, and Giuliano Brancolini with support from the Italian National Antarctic Research Program, the Osservatorio Geofisico Sperimentale (OGS), the Scientific Committee on Antarctic Research, the U.S. Geological Survey (USGS), and the U.S. National Science Foundation. Limited copies of the symposium abstract volume (Terra Antarctica, 1994, 1(2), 251 pp.) are available from Alan Cooper, U.S. Geological Survey, MS 999, 345 Middlefield Road, Menlo Park, CA 94025. Please include your e-mail address, phone, and fax numbers.
Return to Science and
Society 