1 ADCP Primer OC679: Acoustical Oceanography. 2 Outline ► Principles of Operation The Doppler Effect The Doppler Effect BroadBand Doppler Processing.

11

ADCP PrimerADCP Primer

OC679: Acoustical OC679: Acoustical OceanographyOceanography

2

OutlineOutline

► Principles of OperationPrinciples of Operation The Doppler Effect BroadBand Doppler Processing Three-dimensional Current Velocity Vectors Velocity Profile ADCP Data ADCP Pitch, Roll and Heading Ensemble Averaging Echo Intensity and Profiling Range Sound Speed Corrections Measurements Near Surface or Bottom Bottom Tracking

(based on RDI technical tips: http://www.rdinstruments.com/tips/tips.html)

3

OutlineOutline

►Optimizing Your ADCP SetupOptimizing Your ADCP Setup ADCP Setup Parameters Trade-off Triangle:

►Resolution►Range►Random Noise

Operation Modes

(based on RDI technical note: http://www.rdinstruments.com/tips/tips_archive/optimizesetup_1203.html)

4

OutlineOutline

►Practical OperationPractical Operation ADCP Setup with PlanADCP Data collection with VmDas Reprocess real ocean data with VmDas Replay data with WinADCP Data example (SW06)

5

Principles of Operation:Principles of Operation:The Doppler EffectThe Doppler Effect

► Speed of sound = frequency × wavelength: C = fλ

► The Doppler effect is a change in the observed sound pitch that results from relative motion. The Doppler Shift if the difference between the frequency you hear when standing still and what you hear when you move: Fd=Fs(V/C). The Doppler effect measures only relative, radial motion.

► ADCP uses the Doppler effect by transmitting at a fixed frequency and listening to echoes returning from scatterers in the water

6

7

Principles of Operation:Principles of Operation:BroadBand Doppler ProcessingBroadBand Doppler Processing

► If a particle moves away from the transducer, it would take longer time for the second echo to reach the transducer, than for the first. BroadBand ADCPs use phase differences to determine time dilation. The phase differences are exactly proportional to the particle displacement.

► Figure to the right show that Doppler frequency shift and time dilation are equivalent.

► BroadBand ADCPs use time dilation by measuring the change in arrival times from successive pulses

► Long time lags increase precision, but introduce ambiguity problems (B & C upper figure).

► RDI uses autocorrelation techniques for comparing echoes.

8

Principles of Operation:Principles of Operation:Three-dimensional Current Velocity Three-dimensional Current Velocity

VectorsVectors

► Multiple Beams ► from each pair of beams get 1 component of horizontal velocity + 1 component vertical velocity

► assumes current Homogeneity in a Horizontal Layer

► Calculation of Velocity with the Four ADCP Beams

► Error velocity: Why it is useful► really do not need 4 beams to compute U in 3-space► but it provides an estimate of error velocity as the difference in the w estimates► this can bused to detect a) inhomogeneity in measurement or maybe more importantly b)

bad ADCP beam

► The Janus Configuration► Roman god who looks both forward and back

u1 = v sinθ + w cos θ u2 = -v sinθ + w cos θu3 = u sinθ + w cos θ u4 = -u sinθ + w cos θ

9

Principles of Operation:Principles of Operation:Velocity ProfileVelocity Profile

► Depth Cells (bins) and Range Gating

Echoes from far ranges take longer to return to the ADCP than do echoes from close ranges. Profiles are produced by range-gating the echo signal

The velocity is averaged over the depth of the entire depth cell

► The Weight Function for a Depth Cell

The echo from the farthest part of a cell contributes signal only from the leading edge of the transmit pulse. The echo from the closest part of a cell contributes echo only from the trailing edge

As a result the velocity in each depth cell is a weighted average. Also, each depth cell overlaps adjacent depth cell. This overlap causes a correlation between adjacent depth cells of about 15% if transmit pulse is equal to depth cell size

10

Principles of Operation:Principles of Operation:ADCP DataADCP Data

► Velocity Beam, ADCP, Ship & Earth coordinates

► Echo Intensity Receiver Signal Strength (dB)

► Correlation Measure of data quality scaled so that expected

correlation, given high S/N ratio, is 128

► Percent good Variety of rejection criteria (correlation, error velocity,

fish detection)

► Bottom-track Data

11

Principles of Operation:Principles of Operation:ADCP Pitch, Roll and HeadingADCP Pitch, Roll and Heading

► Conversion from ADCP- to Earth Reference (trigonometry & depth cell mapping)

► Measuring ADCP Rotation and Translation Rotation (heading): flux-gate and

gyrocompass Rotation (pitch and roll):

inclinometers & vertical gyro Translation: bottom-tracking, GPS,

reference (“no-motion”) layer

12

Principles of Operation:Principles of Operation:Ensemble AveragingEnsemble Averaging

►ADCP errors: random errors & bias Random errors could be reduced by ensemble

averaging: ~N-1/2

Bias depends on several factors: temperature, mean current speed, signal/noise ratio, beam geometry, etc.

►Averaging inside the ADCP vs. averaging later Conversion of the data prior to averaging Data transmission can slow down ping processing

13

Principles of Operation:Principles of Operation:Echo Intensity and Profiling RangeEcho Intensity and Profiling Range

► Echo intensity:

Sound absorption (exponential decay of echo intensity with increasing range) – increases in proportion to frequency

Beam spreading (range squared) Transmit power (longer pulses put more energy

into the water) Scatterers Bubbles

14

Principles of Operation:Principles of Operation:Sound Speed CorrectionsSound Speed Corrections

►ADCP automatically computes sound speed and corrects velocity based on measured temperature and assumed salinity: Vcorrected=Vuncorrected(Creal/CADCP)

►Depth cell length: Lcorrected=Luncorrected(CADCP/Creal)

15

Principles of Operation:Principles of Operation:Measurements Near Surface or BottomMeasurements Near Surface or Bottom

16

Principles of Operation:Principles of Operation:Bottom TrackingBottom Tracking

► Range to bottom plus three components of bottom velocity. Bottom velocity is used as reference velocity to calculate true current speed (when measured from moving ship). Another application is ice tracking.

► Bottom tracking is implemented using separate pings from water profiling. Requires longer pulses to illuminate bottom completely at one time. Echo is divided in 128 depth cells and ADCP searches through them to find the center of the echo.

17

18

19

20

sound speed and thermoclines

beam angles steep enough

problem arises in assigning depth bins based on a time measurements

21

bottom-tracking

22

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:Setup ParametersSetup Parameters

►Depth range of measurements►Spacing between measurements: in

depth and time►Data averaging►Deployment duration

23

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:Trade-off TriangleTrade-off Triangle

24

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:ResolutionResolution

► Resolution (Depth) vs. Random Noise Doubling depth resolution will double the random

noise, and v.v.

► Resolution: Depth vs. Time Doubling depth resolution without increasing the

random noise will require 4 x the measuring period, and v.v.

► Resolution vs. Deployment Length Doubling the number of depth cells doubles the

power consumption

25

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:RangeRange

► Range vs. Frequency Twice the acoustic frequency will reach about half as far

► Resolution vs. Range Doubling cell size injects more energy into the water –

adding 10% to the profiling range and v.v.

► Range vs. Noise 1.3 x by using narrower bandwidth mode – at cost of 2 x

velocity standard deviation

► Range: external factors Usually, profiling range is enhanced by colder and fresher

water and by more suspended material

26

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:Random NoiseRandom Noise

► Random Noise vs. Resolution Velocity precision improves by

► Random Noise: Dynamic Conditions Velocity precision degrades due to more turbulence,

greater change in velocity across the depth cell, higher boat and water speeds, or greater heave, pitch and roll of the ADCP mounting

► Random Noise vs. Operating Modes Operating modes are separated in two classes: short lag

(1, 12) and long lag (11, 8, 5). Longer lags return more precise velocity data, but have limited profiling range and maximum measuring velocity

No. pings Depth Cell size

27

Optimizing Your ADCP Setup:Optimizing Your ADCP Setup:Operation ModesOperation Modes

Operating Operating ModeMode

11 1212 1111

What it is?What it is? A standard A standard ping mode. All ping mode. All the sensors the sensors are examined are examined each ping.each ping.

Each ping consists of a Each ping consists of a sequence of sub-pings, sequence of sub-pings, which are summed and which are summed and then converted to earth then converted to earth coordinates. Orientation coordinates. Orientation is checked once per is checked once per sequence.sequence.

Transmits hundreds of Transmits hundreds of narrow pulse pairs with pair narrow pulse pairs with pair members separated by a members separated by a wide time lag. Fixes with wide time lag. Fixes with greater time separation greater time separation permit more precise permit more precise velocity estimate.velocity estimate.

Profiling Profiling RangeRange

HighestHighest Mode 1 or less with Mode 1 or less with smaller cellssmaller cells

Limited to a few metersLimited to a few meters

AdvantageAdvantage Higher Speeds Higher Speeds RobustnessRobustness

Most flexible mode. Most flexible mode. Allows higher resolution Allows higher resolution and/or lower std than and/or lower std than mode 1. Lower power mode 1. Lower power consumption.consumption.

High Resolution (1 to a few High Resolution (1 to a few centimeters, depending on centimeters, depending on acoustic frequency)acoustic frequency)

PrecisionPrecision Few cm/sFew cm/s BetterBetter Highest (mm/s)Highest (mm/s)

LimitationsLimitations Not applicable if Not applicable if orientation changes orientation changes significantly during the significantly during the ping sequence.ping sequence.

Limitations of mode 12;Limitations of mode 12;

Slow flows (profiling range Slow flows (profiling range x maximum velocity < 1 x maximum velocity < 1 mm22/s)/s)

28

Data Example from SW06: Data Example from SW06: Workhorse 300kHz vs. Workhorse 1200kHzWorkhorse 300kHz vs. Workhorse 1200kHz

2 2 2 2

3 4 1 2

1 2

3 4

' ' ' '

sin cos sin cos

sin cos sin cos

' ' ' '4 sin cos 4sin cos

u v w u v w

u u w u u w

u u u uu w v w

expand into mean + fluctuations to get Reynolds stresses

Reynolds stresses from ADCP measurements

structure function

for 3D turbulence

Wiles etal 2006

location zseparation r