2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 8
Presentation Time: 3:20 PM

INFLUENCE OF ANTECEDENT CONDITIONS ON CONDUCTIVITY AND SUSPENDED SEDIMENT DISCHARGE IN A KARST SPRING


TANCREDI, Jennifer1, TORAN, Laura1, TARGET, Danielle2, HERMAN, Ellen K.3 and WHITE, William B.3, (1)Dept of Geology, Temple Univ, 1901 N 13th St, Philadelphia, PA 19122-6081, (2)Dept of Geology, Temple Univ, 1901 N 13th St, Philadelphia, PA 19122, (3)Department of Geosciences and Materials Research Institute, Penn State Univ, 210 Materials Research Lab, University Park, PA 16802, jent@temple.edu

The storm response of karst springs involves changes in stage, discharge, conductivity, and suspended sediment. The type of response depends on the stormwater pathways through the karst network and antecedent conditions. We collected a year of storm data from Nolte Spring in southeastern PA using an automatic sampler and sensors that record stage, conductivity, and temperature. The sensors recorded at 20 minute intervals, and the sampler collected samples hourly, triggered by a rise in the water level. Storm samples and monthly samples were analyzed for suspended sediment and ions.

When the antecedent stage at the spring was low (less than 90 cm), storms resulted in a sharp rise in the conductivity up to 100 uS/cm, with one large storm showing a 900 uS/cm rise. This occurs as the older groundwater in the network is pushed out through the spring ahead of the newer, lower-conductivity stormwater. Storms less than 2.5 cm showed little influence on conductivity. When the stage was high (greater than 120 cm), storms showed a more gradual and smaller rise in conductivity, typically around 20 uS/cm. The smaller increase in conductivity at higher water level suggested less contribution from older water and more from younger water that is more dilute.

During drought the suspended sediment load discharged from Nolte Spring was high (30 mg/L). When the stage rose again, sediment loads in monthly samples dropped to 0 to 5 mg/L. Storm sediment collected post-drought by automatic sampler transported 3 to 5 mg/L above the background sediment concentration. The automatic sampler collected mostly small storms of 0.5 to 1.5 cm total precipitation. SEM and XRD analysis is ongoing to characterize the composition of the sediment during storm flows and antecedent conditions. The amount of sediment seemed to be related to degree of flushing before the storm. The timing of the responses may be a function of pathways during recharge, which will be examined in more detail.