Published in American Geophysical Union Abstracts and Programs, San Francisco Meeting, 1999.

The manuscript for this is currently in review at GSA Bulletin

Geomorphic Evidence and Model Support for Build-Up of Subglacial Water Behind a Frozen Margin of the Southern Laurentide Ice Sheet

Cutler, Paul M.1, Mickelson, David M.1, Colgan, Patrick M.2, and Douglas R. MacAyeal 3

1 Geology and Geophysics, University of Wisconsin, Madison, WI 53706, USA

2 Geology, Northeastern University, Boston, MA 02115 , USA

3 Geophysical Sciences, University of Chicago, IL 60637, USA

Ponding of meltwater under an ice sheet is geologically significant for three reasons. First, it may influence ice dynamics and the ice-surface profile. Second, upon drainage, the water may form distinct landforms and deposit large amounts of sediment, and third, lake drainage can impact regional hydrology and even sea level. We examine geomorphic evidence for possibly large lake-drainage events and explore the potential for subglacial ponding using a two-dimensional, time-dependent numerical ice-sheet model.

A laterally extensive unit of clast-supported boulder gravel, with clast intermediate axes of up to 2.4 m, exists in three tunnel-channel fans at the glacial-maximum margin of the Laurentide Ice Sheet (LIS) in Wisconsin. Sedimentologic evidence points to transport of the boulders in a flood. Though poor constraints on paleohydraulic parameters cause an order-of-magnitude range of estimated discharge required to move the boulders, all values within this range represent an unusually large drainage event. Calculated peak flow at one pit is between 700 and 5,000 m3 s-1. At the lower end of the range, three possible water sources exist: an extreme precipitation event, a supraglacial lake, or a subglacial lake. For most of the discharge range, however, the most likely source is a subglacial lake.

Environmental factors favored formation of subglacial lakes behind the southern margin of the LIS in Wisconsin. For example, proglacial permafrost was present as ice advanced to its maximum extent, and the hydrogeologic setting of the ice margin probably allowed only partial drainage as groundwater. Indeed, ice-sheet simulations indicate that the ice may have been frozen to its bed for up to 150 km behind the margin at the glacial maximum, the result of advance over frozen terrain. Furthermore, simulated permafrost depths of tens of meters preclude channelized subglacial drainage. Without such drainage, basal melt water could have accumulated for hundreds of years under this portion of the LIS. Only upon degradation of the subglacial permafrost after the glacial maximum would ponded water have been released.