POLLEN AND SAND IN A SMALL LAKE WITHIN A COASTAL DUNE COMPLEX: A SEARCH FOR PROXIES OF DUNE GROWTH AND MIGRATION

Goshorn Lake is at the eastern margin of a large Lake Michigan coastal dune complex near Saugatuck, Michigan. Marl and sand at the base of sediment cores taken from the lake suggest deposition in streams and shallow wetlands approximately 5700 cal YBP. Sapropels dominate the bulk of the cores and represent deposition in a lake formed by a combination of the flooding of a stream valley during the Nipissing transgression and the damming of the stream's mouth by the coastal dunes. The sizes of peaks in sand concentration in the sapropel (determined by wet sieving after LOI) are typically an order of magnitude larger at the western edge of the lake (immediately adjacent to the dunes) than those from the eastern edge (away from the dunes and adjacent to the stream inlet). This suggests that most of the sand is derived from the dunes. Sand will be transported more readily from the dunes to the lake during periods of active dune growth and migration when large portions of the dunes are relatively free of vegetation. Thus sand peaks in sediment cores may serve as proxies for aeolian activity in the dunes. Vegetation on Lake Michigan sand dunes undergoes a well-established ecological succession beginning with dune grass and herbaceous plants followed by pines and shrubs. The final successional stage is a beech-maple-hemlock forest. Early succession vegetation (grass and pines) should be more prevalent during periods of dune mobility while later succession vegetation (beech, maple and hemlock) should be more prevalent during periods of dune stability, suggesting that pollen from lake cores could be a proxy for dune activity. To test this hypothesis, both sand percentages and pollen composition are being determined on samples from a Livingstone core extracted from the western edge of the lake. Preliminary results show that there is no correlation between the concentration of sand in the core and the relative abundance of grass pollen and that peaks in the relative abundance of pine pollen are not in phase with the peaks in sand concentration. Neither of these pollen types appears to be a good proxy for dune growth and migration. Hemlock pollen shows a steady increase with decreasing depth in the upper portion of the core, but does not correlate with sand abundance.