Home      |     Contact Us      |     Users Site SEARCH:       


Helping Improve Velocity Measurements in the Nation’s Rivers and Streams

The content for this article was contributed by David S. Mueller of the U.S. Geological Survey

Note: Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government

One of the several missions of the United States Geological Survey is to provide reliable, impartial and timely information needed to understand the Nation's water resources. As part of this charge, the agency tracks short-term changes in water flow (over several hours) in rivers and streams. These data are made available on a real-time basis in a number of formats (see, for example, http://water.usgs.gov/waterwatch/).

Among the types of water data collected and disseminated are the flow rates in the Nation's various rivers and streams. These data are collected through a variety of means, but one tool is the Acoustic Doppler Current Profiler (ADCP). An ADCP uses acoustic energy, typically in the range 300–3000 kHz, to measure water velocity throughout most of the water column by measuring the shift in the frequency of the acoustic signals reflected from materials suspended in, and moving with, the water.

Schematic of the Acoustic Doppler Velocimeter (ADV)
Figure 1

Analysis showed that the ADCP was measuring velocities that were biased low in comparison to the measurements near the water surface using another instrument, the Acoustic Doppler Velocimeter (ADV). An ADV measures 3D flow in a small sampling volume located a fixed distance from the probe. The low bias found in the ADCP could be caused by ringing, ping-to-ping interference, or flow disturbance. Flow disturbance was the suspected problem because data analysis and subsequent evaluation of the instrument showed no indications of ringing or ping-to-ping interference.

It is commonly assumed that ADCPs are non-intrusive, but the reality is flow is disturbed around the instrument and measurements in disturbed flow will be erroneous. However, the extent and magnitude of flow disturbance effects are not defined (Figure 1).

Because of the ADCPs' ability to measure flow through a water column from a moving boat, rather than at fixed location, the USGS finds these tools useful. However, because of the known flow disturbance around the ADCP, the USGS needed assurances that the ADCP’s measurements are reliable and useful. Among the methods undertaken to evaluate the ADCP were field data collection, laboratory experiments, tow tank tests, and numerical modeling using FLOW-3D. The latter is the focus of this application note.

The USGS has been a user of FLOW-3D for several years. As a part of an evaluation of the accuracy of the ADCP, the agency believed FLOW-3D would add an important level of comfort and reduce the time to reach a conclusion. For this project, the first step was to calibrate the model against the results of a laboratory experiment.


Flow results in 6 locations
Figure 2


Picture of the free surface where the boom holding the ADCP breaks the water
Figure 3(a)

Simulation of the free surface where the boom holding the ADCP breaks the water
Figure 3(b)

Physical versus numerical results
Figure 4

Simulation showing ownward velocities are much stronger and more concentrated in the vicinity of the ADCP
Figure 5(a)

Simulation showing high downward velocities are concentrated in an area that would be measured by the upstream beam
Figure 5(b)

The laboratory experiment of flow around a SonTek/YSI Acoustic Doppler Profiler was conducted by the University of Illinois. It involved a flume 420 cm long by 30 cm wide by 40 cm deep, with water flowing at a mean velocity of 14.3 cm/s with a 32 cm water depth. The data from the experiment was collected via particle image velocimetry. Figure 2 shows the results for six locations in the flow. The red line represents the results given by the FLOW-3D simulation of the flume, while the circular data points were recorded using a laboratory ADV. At a single location, velocities were captured by particle image velocimetry, this is shown with a blue line.

Satisfied that the FLOW-3D model was reasonably close to the experimental data, the next step was to use FLOW-3D to model the conditions in a case where field data had been obtained using a Teledyne RD Instruments 4-beam ADCP in a canal. The model made use of multiple nested mesh blocks and a user-customized upstream boundary condition.

A visual comparison of the free surface where the boom holding the ADCP breaks the water shows that FLOW-3D was very comparable to the observed free surface (Figures 3(a) and 3(b)).

However, free surface modeling was not the real interest; flow velocities at locations measured by ADCPs were. The flow field from the FLOW-3D model was input into Matlab (which mapped beam paths through the flow field and accounted for the size of the ADCP beam and the bin size) and the velocity that would be measured by an ADCP in the position used in the field data case was computed. The results are presented in Figure 4. The blue line represents the average of 5 field measurements and the red line represents the velocity profile predicted by FLOW-3D.

Figures 5(a) and 5(b) show the strength and spatial extent of the vertical velocity distribution caused by the flow being disturbed by the ADCP. These are plan views with the horizontal coordinates in ft. The circle represents the ADCP. In Figure 5(a), it is clear from the color scale that the downward velocities are much stronger and more concentrated in the vicinity of the ADCP than the recovering upward velocities. Figure 5(b) looks deeper into the water and shows that the vertical velocities decrease in strength, but the high downward velocities are concentrated in an area that would be measured by the upstream beam.

While more modeling is planned to test the effects of the immersion depth of the instrument, water velocity, and deployment type, these early results have given the agency confidence that streamflow measurements made using ADCPs following USGS policy are not biased substantially by flow disturbance.

Acknowledgements: