Paper No. 5
Presentation Time: 9:00 AM

THE IMPACT OF RIVER WATER INTRUSION ON TRACE METAL CONCENTRATIONS IN KARST AQUIFERS: A TALE OF TWO “FLOODS”


BROWN, Amy L.1, MARTIN, Jonathan B.2, SCREATON, Elizabeth3, EZELL, John1 and SPELLMAN, Patricia4, (1)Department of Geological Sciences, University of Florida, 241 Williamson Hall, Gainesville, FL 32611, (2)Department of Geological Sciences, University of Florida, 241 Williamson Hall, P.O. Box 112120, Gainesville, FL 32611-2120, (3)Geological Science, University of Florida, 241 Williamson Hall, PO Box 112120, Gainesville, FL 32611, (4)Department of Civil and Environmental Engineering, Michigan Technological University, 1400 Townsend Dr, Houghton, MI 49931, amy.brown@ufl.edu

The effect of surface water intrusion on trace metal concentrations was evaluated at Madison Blue Spring, a first magnitude spring with a mapped phreatic conduit system that discharges from the upper Floridan aquifer to a dark water river. During two periods of elevated river stage in April 2011 (~1m above baseflow) and March 2012 (~3m above baseflow), tannic river water, with Fe and Mn concentrations several orders of magnitude above spring water concentrations, intruded into the conduits of the spring system. The fraction of river water in the phreatic conduits was tracked using conductivity sensors, and a percent river intrusion was estimated using Cl concentrations of surface samples to constrain a two end-member (river water and groundwater) mixing model. The model was used to determine if Fe and Mn were conservative, mobilized, or sequestered through the course of the events. In April 2011 the maximum river water fraction was 15%, and the river water was detectible in the conduits for ~20 days. In March 2012 river water intruded into the cave for ~8 days, completely displacing groundwater in the conduit system, and was observable in the aquifer for at least 30 days.

A comparison of the 2011 and 2012 events suggests that Fe behavior differs between small and large events. During the 2011 event, excess Fe (greater than calculated from the mixing model) was observed as river water entered the spring system, suggesting mobilization of Fe. In contrast, observed Fe was less than that calculated from the mixing model during the beginning of 2012 event suggesting precipitation under the strongly oxidizing conditions present early in the event. Once river water ceased to flow into the spring on day 8 of the 2012 flood, Fe concentration increased over the model estimate by ~ 0.4 µM, corresponding with a drop in dissolved oxygen likely caused by organic carbon mineralization. In contrast to Fe, Mn appears to be mobilized later in both events; 14 days after river intrusion in 2011 and ~35 days after river intrusion in 2012. In the 2012 event, Mn concentrations remained an order of magnitude higher than normal spring water concentrations 50 days after the event began. Mobilization of Fe and Mn in the subsurface should also impact other trace metal concentrations such as Cu, Pb, and As in the aquifer since Fe and Mn oxides sequester these metals.