Suspended Sediment Transport Revealed by Patterns in Turbidity and Electrical Conductivity Curves in Three Cave Systems in Missouri, USA.

Abstract

In cave streams, the movement of sediment as suspended and bed loads are linked to the occurrence of precipitation events on the surface. During precipitation events, the discharge of the streams in the basin increases. As those surface streams flow into cave streams via sinkholes or fissures, the discharge in cave streams increase. In addition to changes in discharge, the response of the surface and cave streams to precipitation include changes in turbidity and electrical conductivity of the stream water. Changes in turbidity (Tu), electrical conductivity (EC), and discharge (Q) of water in the cave stream can be used to infer sediment movement along cave systems. The types of movement are generally considered direct transfer, when surface sediment is introduced into the cave stream and that same sediment is carried through the entire cave stream during one flood even, resuspension of sediment from the cave stream bed, and deposition of sediment on to the stream bed. The probability of direct transfer is likely controlled by the length of the cave stream (with longer streams having a lesser likelihood of direct transfer except during extreme flood events). Resuspension is probably controlled by seasonality (snow and ice will not move sediment and will not cause an increase in discharge). Deposition is probably more common in longer streams as the sediment introduced from the surface travels through only a portion of the cave before the discharge decreases, allowing the suspended sediment to deposit along the cave stream. Previous researchers developed a method to categorize the movement of fine-grained (suspended) sediment into three classes – direct transfer, resuspension, and deposition. These previous research projects were limited in the number of observations. In this study, the movement of fine-grained material as suspended sediment during rain events were classified based on differences in turbidity, electrical conductivity, and discharge within three cave systems in Missouri, USA. Using previously collected data, 208 flood events were identified and classified from three cave sites. Resuspension of sediments during rain events was observed through all three cave sites for 45% of the flood events; whereas the direct transfer of sediments was observed for 1% of the flood events. Over 50% of flood events observed were too complicated to be place in the established classes. These responses exhibited several curve types during one flood event. The intensity of the flood event and the time between precipitation events did not correlate with the