LC and SMBA Updates

1
LC and SMBA Updates

• Office of Surface Water
• Hydroacoustics Webinar
• January 6 and 9, 2008
• David S. Mueller

2
Overview

• Brief description of loop method
• Common problems using loop method
• Discussion of LC updates
• Compass evaluation
• Bad bottom track warning
• Description of reason for no correction
• Brief description of SMBA
• Discussion of SMBA updates
• Correcting discharge with SMBA
• Using multiple moving-bed tests
• Demonstration of SMBA
• WinRiver II Version 2.04

3
Loop Method
Flow
Distance moved upstream
DUP
VmbDUP/Total Time
1. Establish a starting marker 2. Calibrate
compass 3. Make a transect across and back to the
starting marker 4. Compute the measured mbvel and
the mean channel velocity (vmean).
5. Is a moving bed present? (mbvel 0.04 ft/s
and mbvel / vmean 0.01) 6. Make typical
measurement 7. Compute corrected discharge
4
LC (Loop Correction)

• Automate processing of loop data
• Compute corrected discharge, if a moving-bed is
  detected.
• LC is the recommended method for making loop
  corrections. Manual corrections are discouraged.

5
Common Problems
1) Inadequate compass calibration
2) Invalid bottom track or lost ensembles
3) Misinterpretation of LC results
6
Compass Evaluation

• Velocity direction varies between left to right
  and right to left portions of the loop.
• Loop reversal occurs at the maximum distance made
  good (DMG)

7
Compass Evaluation Algorithm

• Identify the ensembles before and after the
  maximum distance made good.
• Compute the mean flow direction for the two
  halves of the loop.
• Compute the difference in flow direction.
• Compute the standard deviation (SD) of flow
  direction using all ensembles.
• If the difference in flow direction is 5
  degrees, print warning that data indicate a
  potential inaccurate compass.
• HOWEVER, if SD 90 degrees, print warning the
  water velocities are too noisy to evaluate the
  compass.

8
Example of Compass Warning
9
Low Velocity
10
Bottom Track Problems

• The loop method relies on accurate bottom track.
• Invalid bottom track or lost ensembles may result
  in an erroneous loop.
• Invalid bottom track is the result of the ADCP
  being unable to measure a boat speed.
• Lost ensembles are a communications issue that
  may be resolved with changes in data radios,
  serial ports, or laptop computers.
• Warning criteria set at 5.
• NOTE Valid bottom track is most critical during
  speed or heading changes. It is extremely
  important to maintain bottom track at the far
  edge while reversing course.

11
Bad Bottom Track Warning
12
Misinterpretation of LC Results

• Closure error was in downstream not upstream
  direction.
• Error was over 1 of mean velocity, but
  moving-bed velocity was too low.

13
Downstream Closure Error
14
Low Velocity
15
Questions on LC?
16
SMBA

• SMBA was developed to allow analysis of
  stationary moving-bed tests.
• Originally developed for the StreamPro
• Can be used with the Rio Grande
• SMBA was enhanced to provide a correction
  algorithm using one or more stationary moving bed
  tests.
• The corrected discharge is computed by
  distributing the moving-bed velocity correction
  based on the near-bed velocity in each ensemble
  of the transects comprising a measurement.

17
Moving-Bed Velocities
Swimming/Stationary Vmb0.17 ft/s
3-Point/Stationary Vmb0.002 ft/s
18
StreamPro Has a Moving-Bed

• Cant
• Use GPS because you have no compass
• Use the loop method because you have not compass
• Can
• Use section by section if you have the software
• Compute the mid-section method by hand
• Use one or more moving-bed tests to correct
  discharge by hand
• Use one or more moving-bed tests and SMBA to
  correct discharge

19
Rio Grande Has a Moving-Bed

• Use GPS
• Good GPS
• Good compass calibration
• Use the Loop Method
• Accurate bottom track
• Good compass calibration
• Use multiple stationary moving-bed tests and SMBA
  to apply the correction.
• Accurate bottom track
• Use the mid-section or section by section method
• Poor bottom track

20
Bed velocity distribution
Measured Mean
Bed Velocity
Moving bed test
21
Multiple Moving Bed Tests
Bed Velocity
Moving bed test
Moving bed test
Moving bed test
Moving bed test
Moving bed test
22
Distributing the Correction

• Compute velocity components 0.1depth above bed
  for each ensemble using 1/6th power law.
• Distribute the moving-bed velocity based on the
  near-bed velocity.
• Compute the corrected velocities

23
Distributing the Correction cont.

• Compute the corrected measured discharge from the
  correct velocities
• Compute the correct total discharge

NOTE This approach is currently be validated
for stationary moving-bed test.
24
Updates to SMBA

• Corrected problem where SMBA would include a
  negative moving-bed velocity in computing the
  correction from multiple moving-bed tests.
• Corrected multiple moving-bed test approach from
  a mean of the moving-bed tests to a true multiple
  linear regression of the tests.

25
Negative Moving-Bed Fix
Old
New
Negative values were included here. Now they are
set to zero.
26
Change to Linear Regression
Previous Approach
Regression Approach
where a is a regression coefficient from
regession of moving-bed velocities
and near-bed velocities from the
moving-bed tests.
27
Example
Sample Data
Old a0.092
New a0.093
28
SMBA Discharge Correction

• Load and evaluate one or more moving-bed tests
• This step uses the raw data file r.000 or .pd0
• If a moving bed exists click Adjust Q
• Select all transects that comprise the
  measurement
• This step uses ASCII output files that must be
  created from WinRiver or RiverSurveyor
• Review the results and save to a file

29
Moving-Bed Test Validation
30
DEMO SMBA
31
Questions on SMBA?
32
WinRiver II Updates

• WinRiver II 2.03 contained a change in the way
  depth screening was handled.
• The USGS was unaware of the change and failed to
  identify the change during testing.
• The affect of the change was identified by the
  USGS and users in China after the release.
• The change only affected measurements using GPS
  as the reference where there was lost bottom
  tracking.

33
Evolution of Depth Computation

• Traditionally, if any of the four beams processed
  a valid depth the simple average of the valid
  depths was reported and used as the depth for
  that ensemble.
• In WinRiver II 1.0, an option was added to
  changed from a simple average to an inverse
  depth-weighted average. This option is enabled by
  default.
• In version 2.03, TRDI added an additional level
  of screening to the depths and marked depths
  invalid if bottom tracking derived velocities
  were invalid.

34
Effect of Change (GPS Reference)

• Versions Prior to 2.03

Version 2.03
35
Additional Switch in Version 2.04
36
Rules in Version 2.04

• Moving-Boat measurements
• Default Box unchecked
• Depths ARE NOT screened using bottom track
  velocity
• Only affects discharge when using GPS as
  reference
• User can choose to apply screening, if desired
• Section by Section Measurements
• Default Box checked
• Depths ARE screened using bottom track
• Default maintains consistency with StreamPro
  section-by-section software
• User can choose to remove screening, if desired

37
Summary

• Use of the bottom track screening in section by
  section measurements may result in this method
  being unusable in some areas where bottom track
  cannot be maintained. Uncheck if this problem is
  observed.
• We have not observed significant problems with
  erroneous depths where bottom track is not valid.
• Checking or unchecking the box does not affect
  the computed discharge for moving-boat
  measurements unless GGA or VTG is used as the
  reference and invalid bottom track occurs during
  the measurement.

38
Hydroacoustics.usgs.gov

• WinRiver 2.04 is available for download now.
• LC and SMBA are available for download from the
  Hydroacoustics Forum. See instructions on the
  main hydroacoustics web site.

39
Questions on WinRiver II 2.04?