Tracing suspended sediment sources in the Upper Sangamon River Basin using fingerprinting techniques

Abstract:

As the awareness of water pollution, eutrophication and other water related environmental concerns grows, the significance of sediment in the transport of nutrients and contaminants from agricultural areas to streams has received increasing attention. Both the physical and geochemical properties of suspended sediment are strongly controlled by sediment sources. Thus, tracing sources of suspended sediment in watersheds is important for the design of management practices to reduce sediment loads and contributions of sediment-adsorbed nutrients from agricultural areas to streams. However, the contributions of different sediment sources to suspended sediment loads within intensively managed watersheds in the Midwest still remain insufficiently explored.

This study aims to assess the provenance of suspended sediment and the relation between channel morphology and production of suspended sediment in the Upper Sangamon River Basin, Illinois. The 3,690-km² Upper Sangamon River Basin is characterized by low-relief, agricultural lands dominated by row-crop agriculture. Sediment source samples were collected in the Saybrook and Wildcat Slough sub-watersheds from six potential sources: row-crop agriculture, forest, floodplains, river banks, pastures, and grasslands. Event-based suspended sediment samples were collected by in situ suspended sediment samplers and ISCO automatic pump samplers from the streams. A quantitative geochemical fingerprinting technique, combining statistically verified multicomponent signatures and an unmixing model, was employed to estimate the relative contributions of sediment from six potential sources to the suspended sediment loads. Our preliminary results indicate that the majority of suspended sediment is derived from channel banks and forest adjacent to meandering reaches in the downstream portions of the watersheds, while only minor amounts of suspended sediment are derived from upland areas adjacent to channelized rivers in the low-relief Upper Sangamon River Basin. These results suggest that poor connectivity exists between upland areas and channelized reaches of headwater streams in low gradient agricultural watersheds. Further work is needed to determine if this finding is event-dependent with connectivity changing according to event magnitude.