North-Central Section (36th) and Southeastern Section (51st), GSA Joint Annual Meeting (April 3–5, 2002)

LAABS, Benjamin J.C.1, MICKELSON, David M.1, COLGAN, Patrick M.2, WINGUTH, Cornelia1, DARTER, Jessica R.1 and CUTLER, Paul M.3, (1)Geology and Geophysics, Univ of Wisconsin - Madison, 1215 W. Dayton St, Madison, WI 53706, (2)Geology, Northeastern Univ, 360 Huntington Avenue, Boston, MA 02115, (3)National Rsch Council, 2101 Constitution Ave, NW (HA-372), Washington, D.C, 20418, blaabs@geology.wisc.edu

We explore ice sheet and bed interactions using a finite-element numerical flow-line model. The model is driven by climate and incorporates input from a regional geologic database that includes topography, sediment thickness, sediment type, bed type (lithified, unlithified), thickness of sedimentary cover over basement, and flow divergence. The model predicts permafrost distribution and thickness in front of and underneath the glacier, and accounts for calving in lakes and the formation of morainal banks.

Our regional geologic database also contains information about the distribution of common glacial landforms such as drumlins, eskers, moraines, and ice stagnation features. The database includes the glaciated region covered by the southern Laurentide ice sheet, and data are arranged into cells of 7.5 minutes by 7.5 minutes. Because our numerical model is 2-D, most of the data in the geologic database are not used in model input and testing at this time. Until a 3-D numerical model is developed, we will not be able to fully use all of the geologic data.

Our database can provide model input data, but is limited by the coarse spacing in the current 2-D model and the quality of the geologic data and its interpretations. Limitations include the time-transgressive nature of landforms, a lack of complete age control on ice-margin positions, and the possibility of palimpsest landforms being associated with younger ice advances. For example, drumlins located near the late Wisconsin maximum position are used to calculate flow divergence, but it is expected that this value would change with time as the ice advanced and retreated. Drumlins located farther up ice are younger and cannot be used to calculate flow divergence during the late-Wisconsin maximum, but can tell us how divergence changes with time. Despite the availability of high-resolution geologic data on the distribution of glacial landforms, the genesis of these landforms is still problematic. The origins of landforms in the database are not completely understood, so we do not use them as input data in the numerical model. Instead, we use model output to understand the conditions of their formation.