Automatic Real-Time Control of Suspended Sediment Sampling Based Upon High Frequency in situ Measurements of

For estimating suspended sediment concentration (SSC) in rivers, turbidity is potentially a much better predictor than water discharge. Since about 1990, it has been feasible to automatically collect high frequency turbidity data at remote sites using battery-powered turbidity probes that are properly mounted in the river or stream. With sensors calibrated to give a linear response to formazine standards, turbidity and sediment concentration should have a linear correlation close to unity for a given size and composition of suspended particles (Gippel, 1995; Foster et al., 1992). For events of limited duration, the physical properties of the suspended particles probably change very little in most streams. A few (less than 10) data pairs spanning the range of concentrations should be sufficient to reliably establish the relation between SSC and turbidity during such events (Lewis, 1996). This relationship provides a means for accurately estimating sediment loads during storm runoff events. In addition, the detailed turbidity record often contains the signature of sediment inputs to the channel from erosion and mass wasting (Lewis and Eads, 1996). For estimating monthly or annual sediment loads, the relation between SSC and turbidity will vary over time with changes in sediment sources, organic loading, or sensor calibration (Gippel, 1995). Thus, the use of a single curve describing the long-term mean relation will yield greater errors than for short-event estimation. Nevertheless, turbidity is probably more useful than water discharge as a long-term predictor of SSC. If the turbidity-SSC relation is roughly linear, load estimates will be nearly unbiased. In contrast, with sediment rating curves (linear in the logarithms), variance estimation is much more complicated (Gilroy et al., 1990) and such models frequently fit the data poorly and are subject to large errors (Walling an