Miquel Canals¹, Chris D. Clark², José Luis Casamor¹, Roger Urgeles¹, and Verónica Willmott¹

¹Department of Stratigraphy, Paleontology and Marine Geosciences; University of Barcelona; Campus de Pedralbes; E-08028 Barcelona; Spain
²Department of Geography, University of Sheffield; Sheffield, 210 2TN; United Kingdom

Contact: miquel@natura.geo.ub.es

The past existence of a large ice stream off the Gerlache Strait draining into the Western Bransfield Basin is suggested by a convex-upward, elongated sediment body now at a mean water depth of 1000 m. Because of its morphological characteristics, i.e., a set of parallel to subparallel ridges and grooves up to 100 km long, with a width of 14 to 21 km, we called this depositional body a bundle structure (Canals et al., 2000). The bundle, with a core that has been interpreted as formed by basal till, is capped by postglacial sediment smoothening the original groove and ridge topography (Willmott, 2000). The bundle is interpreted as the bed imprint of a paleo-ice stream that drained and expanded the Antarctic Peninsula ice cover during the Last Glacial Maximum (as in previous glacial maxima), and is evidence for rapid ice drainage. This Western Bransfield Basin Ice Stream imprint is possibly the most complete record discovered so far (Stokes and Clark, 2001).

Bundles belong to the family of mega scale glacial lineations, which are landforms composed of drift and produced at the sole of the ice presumably by fast ice flow. Because of their size (tenths of km in length, several hundred meters wide), swath mapping and satellite images are required to observe these mega scale lineations. Since they were first recognised for parts of the bed of the Laurentide Ice Sheet (Clark, 1993), their occurrence has been reported elsewhere in previously glaciated terrains both onland and offshore (Shipp et al., 1999; Canals et al., 2000). The conditions in which they are assumed to be formed makes them extremely useful as an indicator of paleo-ice stream location (Stokes and Clark, 1999). Recently, Clark et al. (2001) and Tulaczyk et al. (in press) hypothesised these features to be carved by keels of ice ploughing through the subglacial sediments. Under flow acceleration and convergence, as it is the case for ice stream onset zones, ice basal roughness elements would experience strain and transform into longitudinally -aligned keels of ice protruding down towards the bed. If, as expected, the strength of such keels is higher than the sedimentary bed underneath, they could plough through the sediments carving elongate grooves. The highest the continuity of the grooves, the fastest the ice flow would be. This mechanism, which still requires validation, could both explain the landforms and the high sediment discharge from ice streams, and does not require that the sediments deform viscously.

Since ice streams are one of the most important controls on ice sheet configuration and stability, the location and behaviour of these features are crucial in any attemp to reconstruct past ice sheets (see review in Stokes and Clark, 2001). Advances in sea floor mapping technology and sub-sea floor imaging make this achievable now, and are starting to generate a promising convergence between marine geologists, glaciologists and modellers, both Antarctic and non-Antarctic. The specific climatic, physiographical and geological conditions existing in the northern Antarctic Peninsula make this region particularly suitable for such a challenge: it is time for "paleo-ice stream hunting".

References
Canals, M. et al., 2000. Geology, 28 (1): 31-34.
Clark, C.D., 1993. Earth Surf. Proc. Landforms, 18: 1-29.
Clark, C.D. et al., 2001. VIth Int. Drumlin Symp., Torun, Poland, June 17-23.
Shipp, S. et al., 1999. GSA Bull., 111 (10): 1486-1516.
Stokes, C.R. and Clark, C.D., 1999. Ann. Glaciol., 28: 67-74.
Stokes, C.R. and Clark, C.D., 2001. Quat. Sc. Rev., 20: 1437-1457.
Tulaczyk, S.M. et al., in press. Quat. Internat.
Willmott, V., 2000. Trab. Fin de Carrera, Fac. Geología, Univ. Barcelona, Spain, 44 p.