2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 208-30
Presentation Time: 4:15 PM


HEERSPINK, Brent P., Department of Geological Sciences, Albion College, 611 E Porter, Albion, MI 49224 and LINCOLN, Timothy N., Department of Geological Sciences, Albion College, 611 E Porter St, Albion, MI 49224, BPH11@albion.edu

The Upper Kalamazoo River is a low-profile, gaining, meandering stream with an adjacent wetland corridor and underlying Late Wisconsinan glacial sediments. Groundwater directly enters the river in three ways: focused flow in springs, marked by clean sand boils; diffuse flow at high rates, marked by cold, 10-140 C, sediment in the summer; and diffuse flow at low rates, marked by warm , 14-200 C , sediment in the summer. River-water nitrate-levels range from 2-8 ppm, depending on stage. Previous studies have shown springs to be consistently higher in nitrate, 15-55 ppm, and areas of diffuse flow at low rates to be consistently lower in nitrate, typically <1 ppm. We believe the high nitrate levels of the springs reflect the values of shallow groundwater derived from the adjacent agricultural uplands; the low values of warm seeps results from nitrate reduction as this water passes slowly through organic- rich river-bed sediment. Springs bypass this reduction. Areas of diffuse flow at high rates are highly variable in nitrate concentrations.

The goal of this study was to determine the cause of this variability though examination of pore-water and sediment at several sites of diffuse flow at high rates. Pore-water was extracted at 10 cm intervals with a narrow gauge stainless steel tube. Cores of the loose sediment were collected by freezing the sediment within a PVC coring tube with an ethanol-dry ice mix in a narrow copper tube inserted into the PVC tube. Water chemistry of pore water extracted from the sediment cores was in good agreement with pore- water sampled in situ. All investigated locations of diffuse flow at high rates with high nitrate concentrations were shown to have sand or sand and gravel layers at depths of 20-30 cm, and nitrate concentrations peak at that horizon. Nitrate-metabolizing bacterial populations were also quantified using fluorescent in situ hybridization, however, no correlation can yet be made between the size of these bacterial communities and the corresponding pore-water nitrate levels. We conclude that the permeable layers of sand and gravel are the conduit for nitrate transport from adjacent upland agricultural land, through the surrounding wetland and to the river. Overpressure in these sand layers may lead in some cases to the sand boils and springs.