"Proceedings, Federal Interagency Workshop,
"Sediment Technology for the 21'st Century,"
St. Petersburg, FL, February 17-19, 1998"

Laboratory Measurements of Suspended Sediment Concentration in Unsteady Flows with an Acoustic Backscatter Profiler

By Marcelo H. García and David M. Admiraal


The measurement of sediment concentration is difficult for any flow. The oldest and most trustworthy method of measuring concentration is to physically sample the flow. Although suction sampling has been used for measuring concentrations in unsteady flows, application of the method is not practical for rapidly changing unsteady flows. There are two reasons for this: first, the intake velocity of the suction sampler should match the velocity of the flow, and controlling intake velocity is difficult (although Bosman et al, 1987, have indicated that use of very high sampling velocities may allow accurate sampling without matching the flow and intake velocities), and second, the concentration varies with time sampling the flow over a time span provides the time average of concentration, but not instantaneous concentrations. Some investigators have overcome the second problem by using a rotating sampler that is in sync with the unsteady flow (Staub et al, 1996). This method is not practical for rapidly changing flows since individual samples are too small.

Acoustic profilers can be used to overcome the temporal problems associated with suction samplers and at the same time are less intrusive in the flow. Acoustic profilers, which have been used since the early 1980s, can gather entire concentration profiles, and can gather them nearly instantaneously. The profiler transmits a sound pulse through water. The sound beam is reflected by particles in the water, and the strength of the return signal is a function of the concentration of sediment in the water. The velocity of the transmitted pulse is known (speed of sound in water), so if the strength of the return signal is measured as a function of time a vertical profile of the sediment concentration is obtained. Due to the nature of backscatter from a random suspension of particles a single acoustic profile is not very meaningful, so a large number of the profiles must be gathered and ensemble averaged to obtain a mean concentration profile.

The measuring volume of the profiler is determined by one of two things, the bandwidth of the output electronics, and the pulse-width of the transducer excitation pulse. The transducer we have has an excitation pulse of 10 mS and an output frequency of 500 kHz, yielding a measuring volume of 0.75 cm. The spatial resolution of the profiler is a result of the frequency of the output signal, and is approximately 3 mm for our transducer. If the pulse width and output circuitry of the transducer are modified, it may be possible to decrease the size of the measuring volume to less than 4 mm and improve the spatial resolution to better than 2 mm. The amount of time it takes to gather a profile depends on the height of the profile. For us the height of the profile is 30 cm so the time it takes is about 0.0004 seconds. The profiler we have can gather profiles at up to 100 times per second. We are gathering them at 10 times per second because of a hard drive speed limitation.

An acoustic profiler can only measure uniform sand grain sizes. This is because the return signal strength is a function of both concentration and sand grain size. The return signal from a low concentration of large sand grains may be the same as the signal from a high concentration of small sand grains. There has been some success in separating concentration and grain size using multiple transducers with differing frequencies (Crawford and Hay, 1993), but the accuracy of the results is substantially reduced.

The profiler was calibrated in a vertical duct for sediment concentrations between 0.1 and 1 percent by mass. Comparison of calibration results and in situ suction samples show that the profiler was working well for constant velocity tests. Suction samples can not be gathered during unsteady flow tests so comparisons with calibration results are not possible for these flows. The performance of the sampler should not be affected by the unsteadiness of the flow and unsteady results were certainly plausible.

The profiler was used to gather concentration profiles in a 10 cm high by 30 cm wide duct with a moveable bed. Two data sets were gathered, one set for each of the sand grain diameters of 0.1 mm and 0.5 mm. Each set of data includes two types of flows, a constant acceleration followed by a constant deceleration, and a constant acceleration followed by a constant velocity. The two types of flows showed that there is a time lag between the flow velocity and sediment concentration profile, and that the time lag increases with distance from the bed.


Bosman, J.J., Van der Velden, E.T.J.M., and Hulsbergen, C.H. (1987) "Sediment concentration measurement by transverse suction." Coastal Engineering, 11, pp. 353-370.

Crawford, A.M., and Hay, A.E. (1993) "Determining suspended sand size and concentration from multifrequency acoustic backscatter." J. of the Acoustical Society of America, 94(6), pp.3312-3324.

Staub, C., Jonsson, I.G., and Svendsen, I.A. (1996) "Sediment suspension in oscillatory flow: measurements of instantaneous concentration at high shear." Coastal Engineering, 27, pp. 67-96.

upWorkshop Contributions

[an error occurred while processing this directive]
[an error occurred while processing this directive]