Estimation of suspended sediment concentration by acoustic scattering: an experimental and theoretical analysis for spherical particles

This study deals with an alternate acoustic calibration method to determine suspended sediment concentration (SSC). Calculations of backscattering from a suspension of spherical particles were compared with a series of experimental measurements neglecting attenuation. Backscattering predictions were obtained using the “form function” description of spherical scattering from modal theory and by fitting according to Beer’s law, which was originally used to measure the transmitted light scattered by a collection of particles, to calculate the acoustic backscattering intensity. Measurements were undertaken in a laboratory using a simple experimental apparatus; an acoustic Doppler velocimeter (ADV) operating at a central frequency of 6 MHz was immersed in a 50-l acrylic tank in which known concentrations of glass microspheres (63- to 90-μm size range) were mixed in freshwater to obtain a homogeneous solution. Measurements expressed in terms of signal to noise ratio (SNR) were taken with different acoustic equipment settings and compared to theoretical predictions. Modal theory form function for elastic glass spheres was used taking into consideration the contributions due to natural vibrations induced by acoustic insonification and because the principal constituent of non-cohesive sediments is quartz. Traditional calibrating procedures were performed; the power of the equipment had an important effect on the results. The preliminary theoretical backscattering predictions were qualitatively satisfactory and suggested that the model can be used to generate the calibration curves for different particle properties. It is feasible to use the proposed model to calibrate the ADV from a reduced number of water samples. From this, it is possible to obtain the characteristics of the particles (shape, density, and size distribution) and so estimate the SSC using the modeled backscattering intensity as a function of appropriate form functions and the number of particles in the ADV sampling volume.