CHARACTERIZATION AND MATHEMATICAL MODELING OF SEDIMENTARY FACIES IN SMALL (~1KM2) LACUSTRINE BASINS, MINNESOTA, USA

Paleolimnologic records of past lake-level changes are important for reconstruction of Quaternary moisture patterns. The variations in sediment grainsize, magnetic properties (susceptibility, ARM/IRM ratios), composition, and macrofossil content preserved in the stratigraphic records of small (&le km2) lake basins are frequently attributed to past changes in lake-level. However, predictive, sequence stratigraphic interpretations of these data are hampered because the spatial variability of these parameters is poorly documented for small basins. Additionally, multiple factors influence these sedimentary properties. For example, evolution of trophic status and/or local wind velocities may cause sediment variation that could be misinterpreted as evidence of lake-level change. Here, we compile a database of over 400 surface sediment samples from 10 lakes of varying trophic and mixing states from across Minnesota in order to document intrabasin spatial variability and formally define littoral, slope, and profundal facies in terms of commonly measured sedimentary parameters. We then use the database to test a mathematical model of sediment grain size variation attributable to changes in the levels of small lakes. The model incorporates wind speed and direction, lake-level and area, and sheltering effects of trees and topography to predict wave characteristics and bottom shear stresses present at any given location within a basin at any given time. Site slope, a coefficient of aquatic vegetation, and sediment density and cohesion are used to evaluate the critical shear stresses required for sediment entrainment. Sediment accumulation is estimated based on a diffusion equation and model output consists of a depth-series of minimum possible grain sizes. Output compares well with observations: characteristics that we find in the littoral, slope, and profundal facies are consistent with the minimum grain-sizes simulated for these environments. Simulated spatial distributions of grain sizes are also consistent with the shallow geophysical facies in ground penetrating radar data, which correspond to the facies of the discretely sampled surface sediment data. Based on these comparisons, the model is a useful tool for testing the plausibility of lake-level scenarios inferred from lake sediment cores.