Post on 19-Jun-2020
transcript
UNOLS Ocean Surveyor ADCPOverview
Julia Hummon, Eric Firing
hummon@hawaii.edu, efiring@hawaii.edu
UNOLS Ocean Surveyor ADCP Overview – p. 1/22
Outline
ADCP: recent history
Installations surveyed
Data and conditions
PerformanceInstallationDepth coverageSide lobesErrors in velocity
Specific problems
Conclusions and Recommendations
UNOLS Ocean Surveyor ADCP Overview – p. 2/22
ADCP: Recent History
4 acoustic beams; interpret Doppler shift
Use heading and position to remove ship motion
1980’s: Narrowband (150KHz)
1990’s: Broadband
2000: Phased Array (150KHz, 75KHz, 38KHz)
Comparison: NB150 vs OS75 (Hummon and Firing,JTECH 2003)
UNOLS Ocean Surveyor ADCP Overview – p. 3/22
Acknowledgements:
Sandy Shor, andScientists and Techs who provided data
This work was supported by NSF grant numberOCE03-09966
This report is online at:http://currents.soest.hawaii.edu/reports
UNOLS Ocean Surveyor ADCP Overview – p. 4/22
Installations Surveyed150KHz 75KHz 38KHz
UNOLS Barnes Endeavor
Walton Smith Knorr
Oceanus
Pelican
Seward Johnson II Seward Johnson I
Wecoma
SEA C. Cramer
R.C. Seamans
USCG Healy
NOAA Sette
LoveBoat Explorer of the Seas Explorer of the Seas
JAMSTEC Kaiyo
UNOLS Ocean Surveyor ADCP Overview – p. 5/22
Data Description
Conditions varied:Acquisition mode (nb or bb, cell size)Water depth, topographic slopeSpeed of ship (many or no stations, transit, ice)Weather conditions
specific comparisonsDepth coverage in different modesBottom return shape from OS vs BB, OS vs NBSame instrument, two different installationsSame instrument and ship, with and without window
UNOLS Ocean Surveyor ADCP Overview – p. 6/22
Performance: Installation
50
100
150
200
284.6 284.8 285 285.2 285.4
5
10
15
20
25
30
35
40
45
50AMPLITUDE
ACG, WITH WINDOW
0
20
40
60
80
100
284.6 284.8 285 285.2 285.4
5
10
15
20
25
30
35
40
45
50PERCENT GOOD
PGOOD, WITH WINDOW
0 50 100 150 200
0
10
20
30
40
50
ACG, WITH WINDOW
ON STATION
UNDERWAY
50
100
150
200
298.6 298.8 299 299.2 299.4
5
10
15
20
25
30
35
40
45
50
55AMPLITUDE
ACG, WITHOUT WINDOW
0
20
40
60
80
100
298.6 298.8 299 299.2 299.4
5
10
15
20
25
30
35
40
45
50
55PERCENT GOOD
PGOOD, WITHOUT WINDOW
Knorr: AGC and Percent Good in two installations
0 50 100 150 200
0
10
20
30
40
50
60
ACG, WITH WINDOW
ON STATION
UNDERWAY
UNOLS Ocean Surveyor ADCP Overview – p. 7/22
Depth coverage summary
best worst(bubbles)
150KHzup to 230m (NB8) 0mup to 200m (BB8)
75KHzup to 750m (NB16, NB8) 0mup to 600m (BB10, BB8) 0m
38KHzup to 1200m (NB16, NB24) 0m
UNOLS Ocean Surveyor ADCP Overview – p. 8/22
Performance: Depth coverage ...
... depends on signal/noise:
Number of scatterersLocation: biological activity variesDiurnal cycle: up at night, deep during dayDepth of scattering layers, some strong
Ambient noiseFaster ship increases ambient noiseFaster ship in heavy weather increases bubbles
UNOLS Ocean Surveyor ADCP Overview – p. 9/22
Performance: Side lobes
AGC bottom return shape shows OS side lobes worse thanNB or BB:
Shows up in all frequencies, (least evident in 38KHz)
Difficult to compare between instruments
Visible as “shoulder” or increased thickness
velocity effect is visible
UNOLS Ocean Surveyor ADCP Overview – p. 10/22
Performance: Errors in velocity
Cross-track error:1 degree heading error at 10kts gives 10cm/s error
Use gyro as the primary heading source (reliable)GPS-aided heading should go into serial port
along-track error:Top bin bias:(up to 1+ m/s) edit out bins or fix installationAnomalous shear:(5-15cm/s) from scattering layers: BE AWAREUnflagged bad bins:(5-15cm/s) EDIT prior to averaging
UNOLS Ocean Surveyor ADCP Overview – p. 11/22
Performance Concerns
All ADCPs:
Bubbles
Ringing
No gps-aided heading
Ocean Surveyor
Increased sidelobe strength (vs NB or BB)
UNOLS Ocean Surveyor ADCP Overview – p. 12/22
Acquisition concerns
Concerns:
Bottom track bug (fixed in firmware version XX?)
Problems with serial inputs:Binary garbage in serial dataBinary message in serial dataToo many unnecessary messages
File names not consistent
Heading comes in through serial port (versus chassis)
Timeserver problems
Thermistors can fail
UNOLS Ocean Surveyor ADCP Overview – p. 13/22
Conclusions (I)
Combination of VmDAS and Ocean Surveyor works
Increased depth range of OS75 and OS38 will be veryvaluable for scientific studies
So far, no ambiguity wraps in velocity
UNOLS Ocean Surveyor ADCP Overview – p. 14/22
Conclusions (II)
Beam side lobesOS sidelobes higher than BB and NBDifficult to compare between instruments(installations)
Velocity errors (all ADCPs)Averaged data can include obvious and subtlebiasesScattering layers do cause a visible effect in velocity
Installation quality is criticalGyro errors induce cross-track velocity errorsRinging causes bias in top binsBubbles are BAD, affecting depth penetration andaccuracy
UNOLS Ocean Surveyor ADCP Overview – p. 15/22
Recommendations
Every installation needs oversight
Designate a central repository of ADCP installationinformation
Deep water installations (frequently out of bottomtrackrange) should have gps-aided heading
Improved automated editing needed
Monitor instrument quality over time and compareinstruments with each other. Need to:
Determine data parameters, site characteristics, testprocedureIdentify locations and manage scheduling
UNOLS Ocean Surveyor ADCP Overview – p. 16/22
Gulf Stream Cross-section
simultaneously logged
NB150 (8m bins) OS75 (BB, 8m) OS75 (NB, 16m)
37.5°N 38°N 38.5°N
0
100
200
300
400
500
600
700 (OSB, ping−edited)
−0.5
0
0.5
1
1.5
2
37.5°N 38°N 38.5°N
0
100
200
300
400
500
600
700 (OSN, ping−edited)
37.5°N 38°N 38.5°N
0
100
200
300
400
500
600
700 (allNB, ping−edited)
met
ers
UNOLS Ocean Surveyor ADCP Overview – p. 17/22
side lobes←−
0 20 40 60 80 100 120 140 160 180
200
220
240
260
280
300
320
340
360
AGC counts
met
ers
Endeavor NB150, OS75
OS75 (8m)
NB150 (8m)
85% of water depth
side lobesdiffer
UNOLS Ocean Surveyor ADCP Overview – p. 18/22
side lobes affect velocity←−←−
40 50 60 70 80 90
200
250
300
350
400
450
500
550
on station
underway
ACG countsde
pth,
met
ers
LIDEX: AGC bump causes "S" in velocity−10 −8 −6 −4 −2
200
250
300
350
400
450
500
550
on station
underway
southward ocean velocity
cm/s
UNOLS Ocean Surveyor ADCP Overview – p. 19/22
heading error←−
170 175 180 185 190 195 200 205 210 215 220
−2°
0°
2°
4°
decimal day
gyro
err
or
gyro error (compared to Seapath)
RB0304, A16N CLIVAR repeat line
170 175 180 185 190 195 200 205 210 215 220−20
−10
0
10
20
30
decimal day
cm/s
cross−track velocity error
UNOLS Ocean Surveyor ADCP Overview – p. 20/22
top bin bias←−
UNOLS Ocean Surveyor ADCP Overview – p. 21/22
edit before averaging←−
360.32 360.34 360.36 360.38 360.4 360.42 360.44 360.46 360.482
3
4
5
6
7
8
9
10averaged data
averaged data
bias in averaged data
no biasm
/s
EX0251 bin 3 abs measured velocity: underway bias towards zero
360.32 360.34 360.36 360.38 360.4 360.42 360.44 360.46 360.480
2
4
6
8
10single−ping data
m/s
single−ping data
bias in raw data
no bias
UNOLS Ocean Surveyor ADCP Overview – p. 22/22