Measuring suspended sediment in sand-bedded rivers using down-looking acoustic doppler current profilers

The use of side-looking acoustic Doppler velocity meters (ADVMs) to estimate fluvial
suspended-sediment concentrations (SSC) has become more operational by the U.S. Geological Survey in recent years; however, direct transfer of these techniques to down-looking acoustic Doppler current profilers (ADCPs) currently is not widely feasible. Key assumptions in the sidelooking ADVM method related to sediment homogeneity within the acoustic measurement volume are almost never met in wide, sand-bedded rivers because SSC and particle size commonly vary with depth and location in the river cross section. The use of ADCPs to estimate SSC has been investigated by researchers, but the requirements and limitations of an operational method that could be successfully applied at many locations are not well defined. If an operational method could be developed, the use of ADCPs, which are routinely used for flow measurements, would revolutionize sediment science by providing rapid measurements of
sediment flux and spatial distribution. We collected detailed datasets in six sand-bedded rivers in the U.S. in 2016-2018, to evaluate the efficacy of using down-looking ADCPs of multiple frequencies to estimate SSC. The datasets included replicate sets of point and depth-integrated suspended-sediment samples and stationary and cross-sectional backscatter profiles using multiple ADCPs with differing frequencies. Reasonable calibrations were developed at all sites measured when calibrating to the coarse fraction (R2 0.66 to 0.98 with slopes close to 0.1 using 1200kHz ADCPs). Calibrations to the fines fraction were problematic because acoustic backscatter response was dominated by coarse particles when present, and substantial attenuation was contributed by coarse particles at some sites. A sensitivity analysis on minimum datasets showed that good calibrations could be developed using two verticals of data collected over a range of backscatter and sediment conditions, with a minimum of three points sampled for sediment within each vertical. Overall, results to date show great promise in using ADCPs to rapidly estimate and visualize SSC with high spatial resolution, and a new beta software tool called Sediment Transect Acoustics simplifies data processing. Improvements are underway to the beta software used in processing to allow incorporation of more acoustic and sediment characteristics and to estimate SSC in areas of the river cross section unmeasured by the ADCP.