Link to USGS home
Measurement of Sediment Load in Agricultural Watersheds Using Acoustic Technology

"Proceedings, Federal Interagency Workshop,
"Sediment Technology for the21'st Century,"
St. Petersburg, FL, February 17-19, 1998"

Measurement of Sediment Load in Agricultural Watersheds Using Acoustic Technology

By Roger A. Kuhnle, Robert W. Derrow II, and Carloas V. Alonso


For many years, the National Sedimentation Laboratory has been a leader in sediment transport research. Research on sediment transport is currently being conducted in laboratory flumes and in Demonstration Erosion Control (DEC) Watersheds. The majority of the sampling effort on the DEC watersheds has been concentrated on the Goodwin Creek Experimental Watershed. Basic concepts in sediment transport developed from experiments conducted under controlled conditions in the laboratory are tested with field data collected on agricultural watersheds. The knowledge gained about sediment transport is incorporated into integrated watershed models (AGNPS, FRAME) which provide our clients with tools needed to make informed management decisions concerning agricultural watersheds.

The high manpower demands imposed by today's total load sampling procedures makes it imperative to explore less expensive methods. To this end, new automatic instruments for the measurement of sediment in transport in laboratory flumes and field streams are being developed in a cooperative project by researchers from the National Sedimentation Laboratory (NSL) and researchers from the National Center for Physical Acoustics (NCPA). Employing acoustics and modern electronics these new instruments will have the potential to lower the costs of collecting sediment transport data and increase the number and coverage of runoff events which are sampled.

Three different acoustic sediment measuring devices are under development at the NSL and NCPA. These are the SedBed Monitor (SBM), the Suspended Sediment Measurement System (SSMS), and the Acoustic Gravel Transport Sensor (AGTS).

Use of a point gage to profile the sediment surface of a flume channel is time consuming at best and under conditions of sediment transport nearly impossible. The SedBed Monitor was developed to accurately measure bed surface transects in a timely manner while the water and sediment are in motion in an experimental flume. The SBM is a computer-controlled, high resolution underwater distance measuring device. It consists of a high frequency (1 MHz), acoustic pulse-echo system which graphically displays the distance from the transducer to the bed surface on a computer screen in real time as well as storing the data to disk. The sampling distance ranges from 0.033 to 3.0 m with a resolution of 0.5 mm. The system can be equipped with up to eight channels to allow multiple data collection sites. Using two sensors, the SBM has been used to measure the rate of bed load transport in a laboratory flume with a uniform-sized sand bed. The collection of two streamwise transects measured within a short time interval allowed the rate of bed load sediment transport to be calculated as the product of the mean migration rate of the bed forms, the mean height of the bed forms, the density of the sediment, and a constant related to the shape of the bed forms. The sediment transport rates calculated from the bed surface transects were very close to mean transport rates measured using an independent method in the flume return pipe. Using bed surface transects to measure sediment transport rates has the advantage of a much shorter period of time necessary to obtain an accurate average transport rate when compared to conventional methods (Kuhnle and Derrow, 1994). Recently a field version of the SBM was developed with 16 sensors, a sealed case for outdoor use and the option of using battery power instead of AC power. It is currently undergoing testing on the Goodwin Creek Experimental Watershed.

An accurate measurement of the rate of suspended sand transport is a difficult task especially in flashy streams. Studies on Goodwin Creek watershed have shown that depth integrated samples over the entire flow depth at several locations across the channel are needed to adequately define the mean sand concentration. A manual sampling system using boom mounted DH-48 samplers has been developed at the NSL to accurately sample the sand in transport. Set up and operation of this system, however, is too expensive to justify use in all but a very few instances. Development of an economic automatic system to accurately measure the mean sand concentration in alluvial channels is the goal of this project. The Suspended Sediment Measurement System (SSMS) is currently under development to meet this need. The SSMS basically functions by sending an acoustic pulse into the water column and measuring the acoustic energy backscattered by the suspended sand in the water. The level of backscattering has been shown to be proportional to the concentration of sand in the water column. The system uses three frequencies (1, 2.25, and 4.3 MHz) to collect information about the size distribution as well as concentration of sediment moving in suspension. The SSMS is currently being tested under controlled conditions in the laboratory (Derrow and Kuhnle, 1996). Tests in a laboratory flume with starved and live bed conditions and tests in a field channel will follow.

Sampling of gravel in transport as bed load is a difficult task. Movement of gravel has been shown to vary drastically in time at a given point and laterally across a cross section at a given time. Boom-mounted modified Helley-Smith (MHS) samplers (Willis et al., 1986) and automatic continuously-recording box samplers (Kuhnle, 1991) have been used successfully to sample bed load on Goodwin Creek. However, sampling with the modified Helley-Smith requires 2 operators to be on site, and the box samplers may fill up before the end of the runoff event. Thus it is difficult to sample the beginning of runoff events with the MHS samples and difficult to sample the end of the event with the box samplers. To solve some of these problems an Acoustic Gravel Transport Sensor (AGTS) is in the conceptual design stage. As presently conceived the AGTS will use an underwater microphone to determine when gravel is in transport by the acoustic emissions from grains as they roll over the bed. Information on the flow strengths necessary for gravel initiation, as well as the sizes and numbers of particles in motion will be measured by this sensor. More information about this project and other NSL research can be found at our web site:


Derrow, R. W. II, and Kuhnle, R. A., Preliminary Analysis of NCPA-NSL Suspended Sediment Calibration System. Proceedings of the 6th Federal Interagency Sedimentation Conference, Las Vegas, Nevada, p. V-58 - V-65, 1996.

Kuhnle, R. A., Bed Load Transport on Two Small Streams. Fifth Federal Interagency Sedimentation Conference, Las Vegas, Nevada, p. 4-139 - 4-146, 1991.

Kuhnle, R. A., and Derrow, R. W. II, Using the SedBed Monitor to Measure Bed Load. Proceedings of Fundamentals and Advancements in Hydraulic Measurements and Experimentation, A.S.C.E., p. 129-138, 1994.

Willis, J.C., Darden, R. W., Bowie, A.J., Sediment transport in Goodwin Creek. Fourth Federal Interagency Sedimentation Conference, Las Vegas, Nevada, p. 4-30 - 4-39, 1986.

Workshop Contributions

[an error occurred while processing this directive]