Hydroacoustic Current Meters for the Measurement of Discharge in Shallow Streams and Rivers

Technical Evaluation at Stennis Space Center

An ADV technical evaluation was conducted at the USGS Stennis Space Center Hydraulics Laboratory on February 1, 2001. SonTek representative Craig Huhta participated in the testing and supplied ADV test data and data analysis, some of which are presented here.  

The purpose of the tests was to evaluate the velocity-measurement accuracy of the ADV in a laboratory setting.  The tests were conducted in the laboratory tow-tank facility.  The tow tank consists of a concrete channel with a 12 ft X 12 ft cross section and length of 450 ft.  An electric tow cart moves along tracks above the tank at user-selectable speeds (in this discussion the terms "velocity" and "speed" are used interchangeably) of 0.05 to 16 ft/sec. 

Tow tank and tow cart

To test the ADV in the tow tank, the tank had to be "seeded" so that there was enough particulate material suspended in the water for the ADV to work properly. The seeding was accomplished by pumping a mixture of water and lime into the tank:

Seeding the tow tank

The ADV was attached to the tow cart so that it was suspended in the tow tank and tested by moving the tow cart at known speeds; ADV-measured water velocities could thus be compared to the tow cart speeds.  The ADV was tested at tow-cart speed ranging from about 0.1 to 5 ft/sec. The ADV was mounted at various yaw angles during the tests:

The following yaw angles (degrees) were tested: zero, positive 10, positive and negative 20, positive and negative 30, and positive and negative 40. 

The ADV puts out velocity data at one-second intervals.  The one-second ADV speeds were averaged over 40 seconds to compute the ADV speed for all runs.  The following table compares ADV to tow-cart speeds for all runs that were made at yaw angle of zero and 10 degrees:

[ADV, acoustic Doppler velocimeter; deg, degrees; ft/s, feet per second; all data are provisional and subject to revision]

The data from the technical evaluation were analyzed by comparing ADV to tow-cart speeds.  Statistics were generated for the departures of ADV from tow-cart speed for the different yaw angles used in the testing, as well as for all runs at all angles.  A summary of the statistics is presented in the table below.  Linear-regression statistics are also presented in the table.  Note that a particular yaw angle is for positive and negative angles; for example, the yaw angle of 40 degrees represents all runs with angles of positive and negative 40 degrees.

[ADV, acoustic Doppler velocimeter; ft/s, feet per second; <, less than; >, greater than; R², correlation coefficient; all data are provisional and subject to revision]

Some results of the testing include:

• For 14 of the 21 runs at 0 and 10 degree yaw angles, ADV speeds were within 1 percent of the corresponding tow cart speeds; for runs above 0.248 ft/s, all ADV speeds were within 1 percent of the tow-cart speeds, except runs at 2.985 ft/s.  Runs at this speed showed a gradually changing speed over the duration of the run -- something that did not occur at other speeds.  Further runs at this speed at all other yaw angles produced similar results.  Vibration of the mounting apparatus has been proposed as a  possible cause.  

• The lower speed (runs at and below 0.248 ft/s) were likely affected by residual currents in the tow tank.  Residual currents have been measured in the tow tank.  For these tests, residual currents were estimated to be about 0.01 ft/s but could have been greater at times.  

• For mounting angles at and greater than 20 degrees, ADV speed departures of about 1 to 20 percent were prevalent due to flow disturbance from the probes.  

• Other than the runs at 2.985 ft/s, no significant bias errors were found in the test data.  

• One-second ADV speeds were examined for a run at a cart speed of 2.188 ft/s.  The mean ADV speed over 40 seconds for this run was 2.203 ft/s (0.7-percent departure from tow-cart speed).  All one-second ADV speeds were within the range of 2.17 to 2.24 ft/s.  The standard deviation of the one-second speeds was 0.018 ft/s (0.8-percent departure from tow-cart speed).  Similar results were shown for other runs (lower speed runs did show the effect of residual currents).  This means that one-second speeds were similar in accuracy to the speeds averaged over 40 seconds.  

Indiana District Hydroacoustics Main Page
Office of Surface Water Hydroacoustic Support Pages

• The use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the United States Government.
• All information herein is provisional and subject to revision.
• Emphasis on the instruments described in these pages does not imply that these are the only instruments that can be used for this purpose. The instruments were selected for the following reasons: the instruments are readily adaptable to this purpose, the instruments are on Federal purchase schedules, and the instrument vendors have voluntarily committed resources to this application. Other instruments will be investigated within the limits of project funding and schedules.

U.S. Department of the Interior
U.S. Geological Survey
5957 Lakeside Boulevard
Indianapolis, Indiana
46278-1996
USA
URL: http://sunrise.er.usgs.gov/hydroacoustics/acoustic_wading_rod_techeval.shtml
Last modified:Monday, 13-Aug-2007 15:58:48 EDT
E-mail questions and comments to webmaster

USGS Indiana Home Page