Southeastern Section - 63rd Annual Meeting (10–11 April 2014)

Paper No. 5
Presentation Time: 9:25 AM

SEDIMENT SOURCE TRACKING USING STABLE ISOTOPES IN A MIXED-USE WATERSHED, SOUTH CAROLINA


MCCARNEY-CASTLE, Kerry, CHILDRESS, Tristan M. and HEATON, Christian Reid, South Carolina Geological Survey, Columbia, SC 29212, childresst@dnr.sc.gov

Studies concerning sediment characteristics and flux in rivers are complicated by a multitude of spatial, temporal, and sampling challenges. Among the challenges a watershed scientist must face is the identification of the source of sediment in a stream. Identifying the key sources of sediment could be critical in order to support management programs and apply best management practices in streams that may be transport-limited, streams with turbidity levels exceeding total maximum daily limits (TMDL), or streams with degraded habitat. In this study, stable isotopic ratios (δC13, δN15) were used for sediment-source fingerprinting in a 220 km2 mixed-use watershed in the South Carolina Piedmont. The Lawsons Fork watershed was selected because of 1) expanding urbanization in the city of Spartanburg; 2) extremely high sediment loads predicted during preliminary 1-d modeling; and 3) thick flood deposits observed in the downstream reach. A regularly monitored cross-section was set up in the downstream reach equipped with a passive sediment sampler, gage height recorder, and sediment tiles. Approximately 7 km upstream of the cross-section, an automatic pump sampler was installed for unattended collection of flood samples alongside an existing USGS real-time gage (02153590). Approximately 100 sediment and soil samples were collected from areas of various land uses within the basin (i.e., forested, urban, pasture) to compare with in-stream (mixed) suspended sediment collected in the downstream reach. The objective was to determine whether the high sediment loads observed in the stream (e.g., 150 NTU) are derived from bank erosion (legacy sediment), overland runoff (pasture vs. urban vs. forested), remobilization of channel material, or transport out of an upstream reservoir. While we expect our results to vary with forthcoming data, our preliminary results indicate that the bulk of sediment in the water column during flood events is derived from the banks of the river, with the composited isotopic ratio indicating a percent frequency occurrence of 87 and 95 percent at the 50th and 99th percentiles, respectively, and a mean frequency occurrence of 86.5%.
Handouts
  • Castle_GSA2014.pptx (16.5 MB)