research article effects of noise and absorption on high...

Research ArticleEffects of Noise and Absorption on High FrequencyMeasurements of Acoustic-Backscatter from Fish

Masahiko Furusawa

Tokyo University of Marine Science and Technology 4-5-7 Kounan Minato Tokyo 108-8477 Japan

Correspondence should be addressed to Masahiko Furusawa frswfineocnnejp

Received 7 June 2015 Accepted 12 October 2015

Academic Editor Renzo Perissinotto

Copyright copy 2015 Masahiko FurusawaThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Quantitative echosounders operating at multiple frequencies (eg 18 38 70 120 200 333 and 710 kHz) are often used to observefish and zooplankton and identify their species At frequencies above 100 kHz the absorption attenuation increases rapidly anddecreases the signal-to-noise ratio (SNR) Also incomplete compensation for the attenuation may result in measurement errorThis paper addresses the effects of the attenuation and noise on high frequency measurements of acoustic backscatter from fish Itis shown that measurements of a fish with target strength of minus40 dB at 200m depth are limited by SNR to frequencies up to about100 kHz Above 100 kHz absorption coefficients must be matched to local environmental conditions

1 Introduction

Quantitative echosounders (QES) or scientific echosoundershave evolved for routine use in acoustic surveys of fisheriesresources [1] Traditionally QES operate at 38 kHz with lowattenuation a high signal-to-noise ratio (SNR) and smallmeasurement errors [2] Recently however QES are oftenoperated at multiple frequencies including higher frequen-cies for example 18 38 70 120 200 333 and 400 kHz [3] 70120 200 333 and 710 kHz [4] and 125 200 455 and 770 kHz[5] to classify and study fish and zooplanktonMeasurementsat the higher frequencies particularly those above 100 kHzare affected by increasingly high and uncertain absorptionattenuation and therefore increasingly low SNR andmeasure-ment accuracy This paper highlights some considerationswhen measuring with high frequencies estimates associatederrors and proposes mitigation measures to obtain preciseand accurate data

The frequency dependence of backscatter provides usefulinformation for fish species identification [6 7] To evaluatethe accuracy that can be achieved by multifrequency QESthe SNR is quantified considering the limitation of transmitpower due to nonlinear phenomenon and acoustic noisefrom various sourcesThen effectivemeasurement ranges areidentified for various frequencies using a modified diagram

of QES performance [2] Also error from absorption attenu-ation at high frequencies is quantified and discussed

2 Materials and Methods

21 Noise There are several sources of acoustic noise receivedby a QES (Figure 1) such as rain wind-induced wave snap-ping shrimp self-vessel and heat (thermal noise) Similar to[2] the noise power spectral level of the vessel self-noise ismodeled here as

119873

119881=

119873

1198810

119891

2 (1)

where 119873

1198810is the noise power spectral level for frequency 119891

projected to 1Hz In the case of the upper vessel self-noise[8] in Figure 1 119873119881

0= 10 log119873

1198810= 155 dB re 1 120583Pa2Hz

(Throughout this paper the MKS system of units is keptin equations but actual values are expressed sometimes inconvenient units as in this case)The lower vessel self-noise inFigure 1 was obtained by fitting (1) to the noise spectral leveldata for a noise-reduced vessel [9] and the value of 119873119881

0is

142 dB The thermal noise spectrum was deduced from theformula in [10] and expressed as

119873

119879= 119873

1198790119891

2

(2)

Hindawi Publishing CorporationInternational Journal of OceanographyVolume 2015 Article ID 589463 11 pageshttpdxdoiorg1011552015589463

2 International Journal of Oceanography

0

10

20

30

40

50

60

70

80

Frequency (kHz)

Rain

Shrimp

VesselVessel

(noise reduced)

Wind

ThermalNoi

se sp

ectr

al le

vel (

dB re

120583Pa

2H

z)

103102101

Figure 1 Noise power spectrum level (dB re 1 120583Pa2Hz) relevant toacoustic surveys of fish resources Heavy rain (100mmh) noise [11]snapping shrimp noise [12] vessel self-noise of RV ldquoHokko-marurdquo(220-tonne steel ship at a speed of 11 knots) [8] vessel self-noiseof RV ldquoSoyo-marurdquo (892-tonne noise-reduced ship at a speed of 10knots) [9] wind-induced noise at rough sea (sea state 6) [13] andsea thermal noise (water temperature 10∘C) [10] are shown

where 10 log119873

1198790= minus75 dB re 1 120583Pa2Hzminus3 Although noise

from rain or shrimp may occasionally dominate the noiselevel for short periods or small regions noise below 100 kHz isgenerally dominated by the vessel self-noise Electrical noisein the QES transducer and receiver may be significant athigher frequencies and will be considered later

22 Signal-to-Noise Ratio The echo level of a single scattereris in general smaller than the echo level of a multiplescatterer and therefore the SNR for single scatterer mustbe considered primarily in the design and utilization ofQES Furthermore the SNR for the average value of volumebackscattering coefficients (119904

119881) as the output of echo integra-

tion decreases especially for single scatterers for the followingreasonThe 119904

119881are obtained for each of themultiple scatterers

single scatterers and even noise in each minimal processingcell called a pixelwhich is the region enclosed by a ping periodand a sampling period along the depth axis [14] The outputof echo integration is the integration of the pixel 119904

119881over an

area of a lager integration cell Let the pixel 119904119881of the echoes

be 119904

119881119864 let the pixel 119904

119881of the noise be 119904

119881119873 let the area which

the echoes occupy in the cell be119860

119864 and let the area of the cell

be 119860

119862 then we have the cell or average 119904

119881

119904

119881119862=

sum 119904

119881119864119860

119864+ 119904

119881119873119860

119862

119860

119862

(3)

Assuming for simplicity that the backscatter within 119860

119862is

homogeneous the SNR for 119904119881119862

is

119877

119862=

119904

119881119864sum119860

119864

119904

119881119873119860

119862

= 119877

sum119860

119864

119860

119862

(4)

where 119877 is the SNR for the echo power Therefore the SNRof the cell 119904

119881(ie 119877

119862) is proportional to the area occupied

by the scatterers divided by the cell area Because 119904

119881119864and119860

119864

are generally smaller for single versus multiple scatterers theworst case 119877

119862is probably that for single scatterers From the

above reasoning the signal for the basis of the SNR should besingle scatterer echoes

In terms of pressure the SNR for a single scatterer is

119877

119865=

119875

119865

2

119875

119873

2 (5)

where 119875

119865is the echo pressure from a scatterer and 119875

119873is

the noise pressure This 119877

119865is expressed by several variables

as described in [2] but the noise expression is slightlychanged as (1) and a transmit power per unit transducerarea is changed from a value obtained by the cavitationthreshold to a value determined to avoid the nonlinear effectof transmission [15 16] From the examples of the frequencytransmit electric power and beam width of QES shown in[3] we adopt 25Wcm2 as the electric power density for allfrequencies examined

The transmitted pressure (1198750) squared is

119875

0

2=

120578119908120587119886

2119863

119868

4120587119903

02

120588119888 =

120588119888119886

2120578119908119863

119868

4119903

02

(6)

where 119908 is the electric power density determined above 120578 isthe electroacoustic efficiency 119886 is the transducer radius 119903

0

is the standard range (1m) 120588 is the density of seawater 119888 isthe sound speed in seawater and 119863

119868is the directivity index

shown approximately by [11]

119863

119868= (

2120587119891

119888

119886)

2

(7)

The single echo pressure (119875119865) squared is

119875

119865

2= 119875

0

2119863

4 119903

0

2

119903

410

02120572(119891)119903120590bs (8)

where 119863 is the directivity coefficient 119903 is the range betweenthe transducer and fish 120572(119891) (dBmminus1) is the frequencydependent absorption coefficient and 120590bs is the backscatter-ing cross sectional area of the scatterer The target strength isexpressed as TS = 10 log(120590bs1199030

2) Using (1) and considering

119863

119868and the bandwidth of the receiving system (Δ119891) the noise

pressure (119875119873) squared is

119875

119873

2=

119873

1198810Δ119891

119891

2119863

119868

(9)

Substituting (6)ndash(9) into (5) we have the expression of theSNR

119877

119865=

4120587

4120588

119888

3sdot

120578119908119863

4120590bs

119873

1198810Δ119891

sdot

119886

6119891

6

119903

410

02120572(119891)119903

(10)

The first factor of (10) is nearly constant the second factormay be assumed to be a constant and the effect of a value

International Journal of Oceanography 3

change can be easily seen by a shift of the SNR scale (from thisreasonwe call the parameters in the second factor ldquoadjustableparametersrdquo) and the last factor greatly changes the SNR Atransducer radius is approximately related to the minus3 dB beamwidth Θ in radian by

119886 cong

051119888

Θ119891

(11)

Substituting this into (10) gives another expression of theSNR

119877

119865= 4 times 051

6120587

4119888

3120588 sdot

120578119908119863

4120590bs

119873

1198810Δ119891

sdot

1

Θ

6119903

410

02120572(119891)119903

(12)

In this expression the factor 119891

6 in (10) disappears and onlythe absorption attenuation remains as a frequency dependentfactor

Since the SNR changes dramatically with 119886 orΘ119891 and 119903it should be an effective way to show the SNR as a function of119891 with parameters 119903 119886 and Θ as in [2] we call the resultantgraph the universal diagram of echosounder performanceThe following values are used for the nearly constant andadjustable parameters

119888 = 1500ms

120588 = 1025 kgm3

120578 = 06 [3]

119908 = 25Wcm2 [3]

119863 = 1

TS = minus40 dB

119873119881

0= 155 dB re 1 120583Pa2Hz

Δ119891 = 25 kHz

(13)

If our objects are only multiple or school echoes as inspecies identification using school echoes we may considerthe SNR for the multiple echo The multiple echo from a fishschool is expressed as

119875

119872

2= 119875

0

2sdot

119903

0

2

119903

410

02120572(119891)119903sdot

119888120591

2

Ψ119903

2sdot 119899120590bs (14)

where 119875

119872is the pressure of multiple echo 120591 is the pulse

duration Ψ is the equivalent beam angle in steradians and119899 is the distribution density of fish or any animals in mminus3Thesecond factor is the propagation attenuation the third factoris the scattering volume and the 1199032 factor included cancels theone-way spreading attenuation in the second factor and thefourth factor is the volume backscattering coefficient 119904

119881 The

Ψ can be approximated by [11] (coefficient is rounded)

Ψ cong

6119888

2

4120587

2119891

2119886

2

(15)

0 200 400 600 8000

100

200

300

400

Frequency (kHz)

Temperature0 to 30 step 5∘

120572(d

Bkm

)

Figure 2 Absorption coefficient by Francois and Garrisonrsquos equa-tions [17 18] as a function of frequency with parameter of watertemperature for constant values of the salinity 35 psu depth 100msound speed 1500ms and pH 8 Note the inversion of temperaturedependence around 70 and 500 kHz

Similarly as in the single echo case the SNR for the multipleecho is derived as

119877

119872=

119875

119872

2

119875

119873

2= 3120587

2120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

119886

4119891

4

119903

210

02120572(119891)119903

= 3 times 051

4120587

2119888

4120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

1

Θ

4119903

210

02120572(119891)119903

(16)

where the first factor may be considered as a constantthe second adjustable parameters and the third importantparameters or variables The adjustable parameter valuesother than 119863 are assumed to be the same as in (13) and thetwo parameter values added are

120591 = 1ms

119899 = 01mminus3 (119878

119881= 10 log 119904

119881= minus50 dB with TS

= minus40 dB)

(17)

23 Absorption Attenuation The absorption coefficient 120572 isa function of water temperature salinity depth and pH [1718] The temperature changes the absorption coefficient mostamong these parameters and the temperature itself changeslargely in time and space For the present purpose thereforeit is sufficient to consider the effect of the frequency andtemperature (Figure 2)

Seasonal changes of the average temperature profiles(Figure 3) at a North Japan Pacific area (42∘N ldquoNorthrdquo orldquoNrdquo) and a South Japan Pacific area (30∘N ldquoSouthrdquo or ldquoSrdquo)were cited from [19] with a slight modification The profileslabeled ldquoNorth1015840rdquo will be explained later From the figure weobserve that there are large seasonal variations and the slopesare steep at depths shallower than 100m and that there arelarge differences between the sea areas


0 100 200 300 4000

5

10

15

20

25

30

Range (m)

North

South

Win

Spr

SumAut

Tem

pera

ture

(∘C)

North998400

Figure 3 Sea temperature profiles for ldquoNorthrdquo (42∘N) ldquoNorth1015840rdquo(41∘N) and ldquoSouthrdquo (30∘N) in the PacificOcean near Japan Originalprofiles in [19] were slightly modified North1015840 profiles are simulatedby the interpolation of North and South profiles

The absorption attenuation is calculated by

119860 (119903) = 2int

119903

0

120572 (119903

1015840) 119889119903

1015840 (18)

It is a common practice in using aQES that we give a constant120572 for each frequency and try to remove this range dependenceof the attenuation Actually however 120572 is not a constant butchanges especially for the water temperature as shown inFigure 2 and expressed in our analysis as

120572 (119903 119891Area Season)

Area = SN1015840N Season = Win Spr SumAut (19)

where Area and Season are expressed by self-evident approx-imate labels

The difference in latitudes of South and North in Figure 3is 12∘ and is too large for our common practice So that wesimulate a more real situation by interpolation and makeprofiles at 41∘N (call ldquoNorth1015840rdquo or ldquoN1015840rdquo) that is 1∘ or 60nautical miles further south than North Simulating ouractual practice of 120572 setting we use as a medium constantvalue the 120572 value at 50m in spring Using N1015840 profiles theattenuation variations at N1015840 from the fixed characteristics atN at six frequencies

Δ119860(119903 119891N1015840 Season) = 2int

119903

0

120572 (119903

1015840 119891N1015840 Season) 119889119903

1015840

minus 2119903120572 (50m 119891N Spr)(20)

are calculatedMoreover let us consider the error caused by using a

fixed value of 120572 assuming that a more suitable temperatureprofilewas givenThedifference of the absorption attenuation

estimates for North and Spring using the temperature profileand the fixed average absorption coefficient is

Δ119860 (119903 119903

119898 119891N Spr) = 2int

119903

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840minus 2119903

sdot

1

119903

119898

int

119903119898

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840

(21)

where the averaging is performed from 0 to 119903

119898 Although

averaging might be performed with weights consideringtarget distribution [20] the simple method is adopted here

3 Results and Discussion

31 Universal Diagrams of Echosounder PerformanceFigure 4 shows the universal diagram for the single echo caseobtained by (10) and (12) the regions where SNR gt 20 dBΘ gt 7∘ and 119886 lt 16 cm are shaded as roughly appropriateregions The SNR border of 20 dB is selected such that thenoise effect can be ignored According to (10) and (12) theSNR curves change +18 dBoctave for 119886 and minus18 dBoctavefor Θ and at low frequencies approximately minus12 dBoctavefor 119903 due to the spreading attenuation The frequency rangesof the roughly appropriate regions are 17ndash400 kHz for119903 = 100m but it shrinks to 20ndash80 kHz for 119903 = 200m It willbe convenient to show rough maximum ranges as a functionof frequency as Figure 5 (the line labeled ldquo120572 errorrdquo will beexplained later)

Figure 6 shows the frequency-SNR curves obtained by(16) for the multiple echo case in a fashion similar to thatof Figure 4 The roughly appropriate region is shaded as inFigure 4 but the lower bound ofΘ is set at 28∘ in order to keepthe bearing resolution angle not too large Differing from thesingle echo case the SNR dependence on 119886 is 12 dBoctavethe dependence on Θ is minus12 dBoctave the dependence on 119903

is relaxed and the roughly appropriate region exists even for400m at near 40 kHz although the decreasing characteristicsof SNR at high frequencies become somewhat severe

Furusawa [2] used the cavitation threshold whichincreases at high frequencies for the power density119908 insteadof the present constant 119908 By the present modification alongwith the slight modification for noise expression (1) the SNRexpression has become more appropriate for quantitativepurposes and the frequency dependence has become simpleand can be clearly understoodWith the presentmodificationthe SNR becomes low and its deterioration becomes severe athigh frequencies but there is no large difference between thetwomethods at low frequencies By the present improvementthe interpretation of the expressions and graphs becomeseasy The constant-119886 SNR curves of (10) increase at lowfrequencies as a function of 119891

6 and decrease sharply athigh frequencies with the increase of absorption attenuationtherefore peaks of the maximum SNR appear and they shiftto a lower frequency with the increase of 119903 The frequencycharacteristics of the constant-Θ curves of (12) are of theabsorption attenuation itself and they are roughly constantat low frequencies where the attenuation is small but theydeteriorate at high frequencies and at large ranges

International Journal of Oceanography 5SN

R (d

B)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

Range 100m Range 400m


Beam width (∘) TD radius (cm)35

7

4

8

16

32

14

28

Frequency (kHz)103102101




Figure 4 Universal diagram of echosounder performance for single fish echoes The SNR for a single fish echo (TS = minus40 dB) is plotted asa function of frequency for parameters of range 119903 transducer radius 119886 and beam width Θ as shown in the 400m range panel Roughlyappropriate regions where SNR gt 20 dB Θ gt 7∘ and 119886 lt 16 cm are shaded

Max

imum

rang

e (m

)

SNR (dB)10

20

30

Frequency (kHz)103

103

102

102

101101

120572 error

Figure 5 The range where the SNR for single fish echo is 10 20 and 30 dB as an index of maximum measurable range The line labeled ldquo120572errorrdquo is the range where absorption attenuation error is estimated to be 1 dB






SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)



Figure 6 Universal diagram of echosounder performance for multiple echoes The SNR for a multiple echo (TS = minus40 dB 119899 = 01 fishm3and 119878

119881= minus50 dB) is plotted as a function of frequency for parameters of range 119903 transducer radius 119886 and beam widthΘ as shown in 400m

range panel Roughly appropriate regions are shaded

The values of adjustable parameters in (13) and (17)are approximate and they are changeable to fit userrsquos ownsounder parameters In such adjustment it is sufficient onlyto shift the SNR scale For example if a vessel is more silentand 119873119881

0is 142 dB electric power density 119908 is 3Wcm2 and

efficiency 120578 is 075 then from the computation

10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)

the SNR increases by about 15 dB and the SNR scales inFigures 4 and 6 may be read adding by 15 dB

Figures 4 and 6 can be applied not only to QES but alsoto fisheries echosounders (FES) An FES operating atmultiplefrequencies usually uses a broader beam for a lower frequencyand the low frequency is used to search a broader area anda high frequency is used to observe fish in high resolutionFor example let us compare the performance of an FES witha combination of 120 kHz and 7∘ beam and 38 kHz and 28∘beam as seen from Figure 4 the SNR for 120 kHz is larger up

to 400m than 38 kHz The beam widths of QES are usuallyselected to be the same at multiple frequencies mainly forthe purpose to observe frequency difference of scattering [3]therefore the SNR at a lower frequency is always higherCommon sense for FES therefore is not always true for QESand caution must be exercised

The discussion in Section 21 suggests that we shouldconsider mainly the self-vessel sailing noise in fish scatteringmeasurements from a cruising vessel In fact Nishimura [8]observed the fact that received noise increased by about 20 dBwhen a ship started tomove Also Korneliussen [21] observedpropeller noise up to 30 dB higher than noise while driftingespecially in shallow waters and above dense and large fishschools These facts are the simple result that we measuresignals on the order of milliwatts near an engine with poweron the order of megawatts [22] If vessel noise is reducedby several countermeasures [23] a noise spectrum level canbe considerably reduced as shown by the lower vessel noiseline in Figure 1 [9] but the vessel self-noise is still generallylarger than the wind-induced noise Of course in cases ofa drifting ship stopping a ship engine using a suspended


minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB


Figure 7The effect of sea thermal noise and receiver noise on SNRfor a target with 119879119878 = minus40 dB at 119903 = 100m and a beam width 7∘The SNR based on all noises (self-vessel sea thermal and receivernoises) is shown by the solid lines for two types of ship with 119873119881

0=

142 dB (noise reduced) and 155 dB The dashed curves indicate theSNRbased only on the self-vessel noise the lower dashed curve is thesame as the corresponding curve in the upper left panel of Figure 4

0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South

Figure 8 Absorption attenuation at 38 kHz and 200 kHz for Northand South seas and four seasons computed by using the temperatureprofiles shown in Figure 3 Season difference is hardly discernible at38 kHz and also between Summer and Autumn at 200 kHz

echosounder from a ship measurement from a buoy or usinga stationary system wind-induced noise will dominate evenin such cases a modification of 119873119881

0will suffice to estimate

the SNR using (10) (12) and (16)

32 SNR at Especially High Frequencies At especially highfrequencies however since the vessel self-noise and wind-induced noise decrease we may have a situation where

the thermal noise and the receiver noise of QES itself aredominant According to [24] the total noise power spectrumlevel including the thermal noise of seawater and hydrophoneand the receiver noise is shown as

119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)

where the first term represents ambient noise other than thesea thermal noise the second term the thermal noise of thesea and hydrophone and the third term the receiver noisewhere 119873

119867is the value determined by hydrophone specifica-

tions and 119865 is the noise figure of a preamplifierThe last termcan be expressed as119873

1198670119891

2 taking out frequency dependenceand assuming 10 log119865 = 5 dB and typical hydrophoneparameters we have 10 log119873

1198670= minus70 dB re 1 120583Pa2Hzminus3

this value is 5 dB larger than the case for the sea thermal noise(2) The SNR for a single echo is derived in a fashion similarto (12) but using the above noise characteristics and 119873

119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)

Figure 7 compares SNR calculated for the two vessel noises bythis equation with the SNR for which only vessel self-noiseis used as in Figure 4 As frequency increases the differenceof the SNR for the two vessels becomes small mainly dueto the contribution from the receiver noise and the effect ofthis noise becomes prominent from sim200 kHz for the noise-reduced vessel with 119873119881

0= 142 dB and sim400 kHz for the

vessel with 119873119881

0= 155 dB To make the receiver noise small

we need a highly efficient transducer an amplifierwith a smallnoise figure and appropriate matching between the amplifierand transducer

33 Countermeasures for Noise Noise removal methods havebeen reported in [21 25 26] Reasonable correction shouldbe possible when SNR is high as gt10 dB but correctionmight be difficult in a low SNR condition when correctionis especially needed Therefore it is important to build ameasurement system including a survey vessel with highSNR (eg to select low frequencies) to conduct survey ingood SNR conditions (eg lowering ship speed) and to selectobject fish (school) echoes in postprocessing (to make theratio sum119860

119864119860

119862in (4) large)

As described in Simrad EK60 Reference Manual [27] aneffective way of monitoring SNR and knowing the maximummeasurable range is to show an echogram of ldquo40 log 119903rdquo TVGoutput whose minimum display level is the minimum TSvalue to be measured minus an expected SNR value andto recognize the range where noise starts to appear as themaximum range For example if the minimum TS is minus60 dBand the expected SNR is 20 dB the minimum display level ofTS should be minus60 minus 20 = minus80 dB


0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz

Figure 9 Deviations of two-way absorption attenuations (Δ119860) at North1015840 sea for six frequencies from the attenuation computed by using afixed absorption coefficient of North Spring and 50m depthThe line labels are shown in 800 kHz panel The dotted vertical lines are drawnat the rough maximum ranges with 20 dB SNR shown in Figure 4

Higher frequencies up to about 800 kHz are mainlyused for fish species classification or discrimination betweensmall animals and fish and ordinarily resultant volumebackscattering strengths are not directly used for abundanceestimation therefore small errors may be admissible Theuppermost frequencies with which this paper deals are lessthan 1MHz and the main object is fish which are the targetsof fisheries and therefore the frequency characteristics of TSare roughly assumed to be flat If our main objects were smallanimals such as zooplankton the frequency characteristics ofTS should also largely affect the SNR Liu et al [28] used ahigh-pass prolate-spheroid liquid model for TS and obtaineduniversal diagrams as shown in Figure 4 but with anotherparameter of body sizeThe results demonstrate that the SNRbecomesmaximum around 1MHz for range 50m and animalsize 1ndash3mm

34 Compensation Errors of Absorption Attenuation First letus see the synoptic characteristics of absorption attenuationFigure 8 shows the absorption attenuation (18) at 38 and

200 kHz using the temperature profiles shown in Figure 3We see that the attenuation is approximately one order ofmagnitude larger at 200 kHz than at 38 kHz that seasondifference is hardly discernible at 38 kHz and also betweenSummer and Autumn at 200 kHz that large difference existsbetween North and South at 200 kHz and that the attenua-tion is larger at North than South at 38 kHz but the inverseis true at 200 kHz the last fact can be explained by Figure 2which shows inversion of the temperature dependence at near70 kHz This figure simply suggests that the correction of theabsorption must be made carefully to match environmentalconditions especially for high frequencies

Figure 9 shows the errorΔ119860 (20) caused by using the fixedvalue of 120572 Variations which are not negligible appear from120 kHz and at 400 kHz the variations become somewhatsmaller but at 800 kHz they become very large The peculiarphenomenon at 400 kHz can be explained by the character-istics shown in Figure 2 The curves for spring correspondto the case that there were only areal difference but notime difference in the temperature profiles as expectedthe variations are rather small As a whole the error may


0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)

Figure 10 Deviations of two-way absorption attenuations (Δ119860) by using fixed values of absorption coefficient averaged over range 119903

119898 The

profile used is of Spring at North shown in Figure 3

be a few decibels at frequencies higher than 120 kHz ifthe temperature profiles are not measured frequently andaccurate absorption coefficients are not given

Figure 10 indicates the errors for more appropriate cor-rection (21) The result for 38 kHz is very small and theresult for 800 kHz is too large and they are not shownAgain 400 kHz shows surprisingly small error The similarcalculation forNorth andAutumn gave similar results and forWinter negligible error but for Summer the error becomesconsiderably larger (not shown) South profiles gave similarcharacteristics From these results if we set an averagingrange 119903

119898 properly for target fish the error by using a

fixed coefficient can be made small However caution mustbe paid to the fact that the above calculation assumed theexact temperature profiles which are not the case in actualpractice It seems better to set the averaging range 119903

119898 at

approximately half of the target range 119903 In place of usingan average coefficient a simpler alternative way is using afixed coefficient at a medium depth as was adopted in thederivation of Figure 9

The area season and method of correction adoptedas above are only examples to show the degree of error

therefore actual errors may be larger or smaller according toeachmethod or customof the user of aQES To evaluate errorit is necessary that each user of QES compute variation in asimilar fashion to match hisher QES sea area and surveymethod

Since the expression of the absorption coefficient byFrancois and Garrison [17 18] so far used in this paper wasderived through theory and many and diverse experimentaldata in various seas its reliability seems to be high andtheir expression is appropriate for our fisheries acousticsapplication But the degree of error claimed in their paperis around 5 and then there is the possibility to introduceerror which is not negligible in attenuation compensationespecially at high frequencies where 120572 value is large Onaet al [29] pointed out that the difference of 120572 values by theexpression of Francois andGarrison and Fisher and Simmons[30] amounts to 4 dBkm at 333 kHz and this value is alsoabout 5 The dashed line labeled ldquo120572 errorrdquo in Figure 5 isthe range 119903 obtained from 2 times 005 times 120572(119891)119903 = 1 where120572 is calculated by Francois and Garrisonrsquos expression thatis the range where 5 error of 120572 gives 1 dB error Therange is 120m at 300 kHz and this range is nearly equal to


the roughmaximum range for 20 dB SNR At higher frequen-cies the120572 error becomes a stronger constraint than SNRThisproblem is not easy to overcome but a possible method isto check the absorption attenuation compensation by the TSmeasurement by the split beam method of standard targetssuspended at different depths

At the frequency regions around 70 kHz and 400 kHzwhere the temperature characteristics of 120572 is reversed(Figure 2) the variation of 120572 is comparatively small Thefrequency of 70 kHz is a good choice for QES next to 38 kHzbecause of the above reason and also because a high SNR ispossible (Figure 4) At 38 kHz use of a fixed 120572 value causesa very small error These frequencies therefore provide arobust QES

4 Conclusions

In order to obtain species identification information fromthe relative frequency response the measurement error mustbe very small for example smaller than 05 dB The errorsconsidered in this paper are added to the calibration error ofsim4 or 02 dB [3 31] and therefore they must be minimal

Since error becomes large when SNR is small it isnecessary to know the limits ofmeasurement conditions fromthe universal diagram as shown in Figure 4

Serious error source at high frequencies is an inappro-priate setting of the absorption coefficient Frequent mea-surements of temperature profiles and adaptive setting of 120572

are necessary The best way is that we set an opportunetemperature profile into an echosounder and it calculates anabsorption coefficient profile to compensate for the absorp-tion attenuation automatically

Both the decrease of SNR and the error caused by absorp-tion attenuation limit the maximum measureable range andto increase the range is fundamentally difficult therefore acountermeasure is to reduce the range to the target fish usingfor example a suspension probe from a vessel [32 33]

Since it is considerably difficult to accurately measurescattering of fish deeper than 200m by a frequency above100 kHz with an error smaller than 1 dB we must be verycareful in high frequency measurements of fish scattering

Conflict of Interests

The author declares that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

K Sawada National Research Institute of Fisheries Engineer-ing and K Amakasu Tokyo University of Marine Scienceand Technology are thanked for their useful comments to thepaper The author would like to thank N Williamson AlaskaFisheries Science Center NOAA USA for his valuablecomments and correction of English Anonymous refereesare thanked for their helpful comments

References

[1] J Simmonds andDMacLennan Fisheries AcousticsTheory andPractice Blackwell Oxford UK 2nd edition 2005

[2] M Furusawa ldquoDesigning quantitative echo soundersrdquo TheJournal of the Acoustical Society of America vol 90 no 1 pp26ndash36 1991

[3] R J Korneliussen N Diner E Ona L Berger and P GFernandes ldquoProposals for the collection of multifrequencyacoustic datardquo ICES Journal of Marine Science vol 65 no 6 pp982ndash994 2008

[4] A M Kaltenberg and K J Benoit-Bird ldquoIntra-patch clusteringin mysid swarms revealed through multifrequency acousticsrdquoICES Journal of Marine Science vol 70 no 4 pp 883ndash891 2013

[5] D Lemon G Borstad L Brown and P Johnston ldquoMooredmultiple-frequency sonar for long-term continuous observa-tions of zooplankton and fishrdquo Journal of Marine AcousticsSociety of Japan vol 40 no 1 pp 37ndash41 2013

[6] R J Korneliussen and E Ona ldquoAn operational system forprocessing and visualizingmulti-frequency acoustic datardquo ICESJournal of Marine Science vol 59 no 2 pp 293ndash313 2002

[7] ldquoThe SIMFAMI project species identification methods fromacoustic multifrequency informationrdquo Final Report to the ECQ5RS-2001-02054 2006

[8] M Nishimura Study on the optimum frequency of echosounders [PhD thesis] Tohoku University Sendai Japan 1969(Japanese)

[9] Y Takao K Sawada YMiyanohana TOkumuraM Furusawaand D Hwang ldquoNoise characteristics of Japanese fisheriesrsquoresearch vesselsrdquo The Journal of the Acoustical Society of Amer-ica vol 103 no 5 p 3036 1998

[10] R H Mellen ldquoThe thermal-noise limit in the detection ofunderwater acoustic signalsrdquo The Journal of the AcousticalSociety of America vol 24 no 5 pp 478ndash480 1952

[11] HMedwin and C S Clay Fundamentals of Acoustical Oceanog-raphy Academic Press San Diego Calif USA 1998

[12] V O Knudsen R S Alford and J W Emling ldquoUnderwaterambient noiserdquo Journal of Marine Research vol 7 pp 410ndash4291948

[13] G M Wenz ldquoAcoustic ambient noise in the ocean spectra andsourcesrdquo The Journal of the Acoustical Society of America vol34 no 12 pp 1936ndash1956 1962

[14] M Furusawa ldquoSection and total scattering strengths of individ-ual fish schoolrdquo Journal ofMarine Acoustics Society of Japan vol38 no 4 pp 177ndash194 2011

[15] F E Tichy H Solli and H Klaveness ldquoNon-linear effects in a200-kHz sound beam and the consequences for target-strengthmeasurementrdquo ICES Journal ofMarine Science vol 60 no 3 pp571ndash574 2003

[16] A Pedersen Effects of nonlinear sound propagation in fisheriesresearch [Doctoral thesis] University of Bergen Bergen Nor-way 2006

[17] R E Francois and G R Garrison ldquoSound absorption based onocean measurements Part I pure water and magnesium sulfatecontributionsrdquoThe Journal of the Acoustical Society of Americavol 72 no 3 pp 896ndash907 1982

[18] R E Francois and G R Garrison ldquoSound absorption basedon ocean measurements Part II Boric acid contribution andequation for total absorptionrdquo Journal of the Acoustical Societyof America vol 72 no 6 pp 1879ndash1890 1982


[19] Marine Acoustics Society of Japan Fundamentals and Appli-cations of Marine Acoustics Seizando Tokyo Japan 2004(Japanese)

[20] D A Demer ldquoAn estimate of error for the CCAMLR 2000survey estimate of krill biomassrdquo Deep Sea Research Part IITopical Studies in Oceanography vol 51 no 12-13 pp 1237ndash12512004

[21] R J Korneliussen ldquoMeasurement and removal of echo integra-tion noiserdquo ICES Journal of Marine Science vol 57 no 4 pp1204ndash1217 2000

[22] R B Mitson ldquoShip noise related to fisheries researchrdquo Proceed-ings of the Institute of Acoustics vol 11 part 3 pp 61ndash67 1989

[23] R B Mitson Ed Underwater Noise of Research Vessels Reviewand Recommendations ICES Cooperative Research Report No209 ICES Copenhagen Denmark 1995

[24] J W Young ldquoOptimization of acoustic receiver noise perfor-mancerdquoThe Journal of the Acoustical Society of America vol 61no 6 pp 1471ndash1476 1977

[25] J L Watkins and A S Brierley ldquoA post-processing techniqueto remove background noise from echo integration datardquo ICESJournal of Marine Science vol 53 no 2 pp 339ndash344 1996

[26] A De Robertis and I Higginbottom ldquoA post-processingtechnique to estimate the signal-to-noise ratio and removeechosounder background noiserdquo ICES Journal of Marine Sci-ence vol 64 no 6 pp 1282ndash1291 2007

[27] A S Kongsberg Maritime Simrad EK60 Reference Manual2012

[28] X Liu M Furusawa E Hamada and C Aoyama ldquoDesigningacoustic transmitting-receiving system for volume backscatter-ing measurement of zooplanktonrdquo Fisheries Science vol 65 no3 pp 410ndash419 1999

[29] E Ona G Macaulay R Korneliussen and D Chu ldquoThe acous-tic absorption coefficient at 333 kHzrdquo Report Working Groupon Fisheries Acoustics Science and Technology (WGFAST)Brest France 2012

[30] F H Fisher andV P Simmons ldquoSound absorption in sea waterrdquoThe Journal of the Acoustical Society of America vol 62 no 3 pp558ndash564 1977

[31] K G Foote H P Knudsen G Vestnes D N MacLennanand E J Simmonds ldquoCalibration of acoustic instruments forfish density estimation a practical guiderdquo ICES CooperativeResearch Report 144 ICES Copenhagen Denmark 1987

[32] E Ona ldquoAn expanded target-strength relationship for herringrdquoICES Journal ofMarine Science vol 60 no 3 pp 493ndash499 2003

[33] K Sawada H Takahashi K Abe T Ichii K Watanabe and YTakao ldquoTarget-strength length and tilt-angle measurements ofPacific saury (Cololabis saira) and Japanese anchovy (Engraulisjaponicus) using an acoustic-optical systemrdquo ICES Journal ofMarine Science vol 66 no 6 pp 1212ndash1218 2009

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Geological ResearchJournal of


Geology Advances in


0

10

20

30

40

50

60

70

80

Frequency (kHz)

Rain

Shrimp

VesselVessel

(noise reduced)

Wind

ThermalNoi

se sp

ectr

al le

vel (

dB re

120583Pa

2H

z)

103102101

Figure 1 Noise power spectrum level (dB re 1 120583Pa2Hz) relevant toacoustic surveys of fish resources Heavy rain (100mmh) noise [11]snapping shrimp noise [12] vessel self-noise of RV ldquoHokko-marurdquo(220-tonne steel ship at a speed of 11 knots) [8] vessel self-noiseof RV ldquoSoyo-marurdquo (892-tonne noise-reduced ship at a speed of 10knots) [9] wind-induced noise at rough sea (sea state 6) [13] andsea thermal noise (water temperature 10∘C) [10] are shown

where 10 log119873

1198790= minus75 dB re 1 120583Pa2Hzminus3 Although noise

from rain or shrimp may occasionally dominate the noiselevel for short periods or small regions noise below 100 kHz isgenerally dominated by the vessel self-noise Electrical noisein the QES transducer and receiver may be significant athigher frequencies and will be considered later

22 Signal-to-Noise Ratio The echo level of a single scattereris in general smaller than the echo level of a multiplescatterer and therefore the SNR for single scatterer mustbe considered primarily in the design and utilization ofQES Furthermore the SNR for the average value of volumebackscattering coefficients (119904

119881) as the output of echo integra-

tion decreases especially for single scatterers for the followingreasonThe 119904

119881are obtained for each of themultiple scatterers

single scatterers and even noise in each minimal processingcell called a pixelwhich is the region enclosed by a ping periodand a sampling period along the depth axis [14] The outputof echo integration is the integration of the pixel 119904

119881over an

area of a lager integration cell Let the pixel 119904119881of the echoes

be 119904

119881119864 let the pixel 119904

119881of the noise be 119904

119881119873 let the area which

the echoes occupy in the cell be119860

119864 and let the area of the cell

be 119860

119862 then we have the cell or average 119904

119881

119904

119881119862=

sum 119904

119881119864119860

119864+ 119904

119881119873119860

119862

119860

119862

(3)

Assuming for simplicity that the backscatter within 119860

119862is

homogeneous the SNR for 119904119881119862

is

119877

119862=

119904

119881119864sum119860

119864

119904

119881119873119860

119862

= 119877

sum119860

119864

119860

119862

(4)

where 119877 is the SNR for the echo power Therefore the SNRof the cell 119904

119881(ie 119877

119862) is proportional to the area occupied

by the scatterers divided by the cell area Because 119904

119881119864and119860

119864

are generally smaller for single versus multiple scatterers theworst case 119877

119862is probably that for single scatterers From the

above reasoning the signal for the basis of the SNR should besingle scatterer echoes

In terms of pressure the SNR for a single scatterer is

119877

119865=

119875

119865

2

119875

119873

2 (5)

where 119875

119865is the echo pressure from a scatterer and 119875

119873is

the noise pressure This 119877

119865is expressed by several variables

as described in [2] but the noise expression is slightlychanged as (1) and a transmit power per unit transducerarea is changed from a value obtained by the cavitationthreshold to a value determined to avoid the nonlinear effectof transmission [15 16] From the examples of the frequencytransmit electric power and beam width of QES shown in[3] we adopt 25Wcm2 as the electric power density for allfrequencies examined

The transmitted pressure (1198750) squared is

119875

0

2=

120578119908120587119886

2119863

119868

4120587119903

02

120588119888 =

120588119888119886

2120578119908119863

119868

4119903

02

(6)

where 119908 is the electric power density determined above 120578 isthe electroacoustic efficiency 119886 is the transducer radius 119903

0

is the standard range (1m) 120588 is the density of seawater 119888 isthe sound speed in seawater and 119863

119868is the directivity index

shown approximately by [11]

119863

119868= (

2120587119891

119888

119886)

2

(7)

The single echo pressure (119875119865) squared is

119875

119865

2= 119875

0

2119863

4 119903

0

2

119903

410

02120572(119891)119903120590bs (8)

where 119863 is the directivity coefficient 119903 is the range betweenthe transducer and fish 120572(119891) (dBmminus1) is the frequencydependent absorption coefficient and 120590bs is the backscatter-ing cross sectional area of the scatterer The target strength isexpressed as TS = 10 log(120590bs1199030

2) Using (1) and considering

119863

119868and the bandwidth of the receiving system (Δ119891) the noise

pressure (119875119873) squared is

119875

119873

2=

119873

1198810Δ119891

119891

2119863

119868

(9)

Substituting (6)ndash(9) into (5) we have the expression of theSNR

119877

119865=

4120587

4120588

119888

3sdot

120578119908119863

4120590bs

119873

1198810Δ119891

sdot

119886

6119891

6

119903

410

02120572(119891)119903

(10)

The first factor of (10) is nearly constant the second factormay be assumed to be a constant and the effect of a value



119886 cong

051119888

Θ119891

(11)


119877

119865= 4 times 051

6120587

4119888

3120588 sdot

120578119908119863

4120590bs

119873

1198810Δ119891

sdot

1

Θ

6119903

410

02120572(119891)119903

(12)




119888 = 1500ms

120588 = 1025 kgm3

120578 = 06 [3]

119908 = 25Wcm2 [3]

119863 = 1

TS = minus40 dB

119873119881

0= 155 dB re 1 120583Pa2Hz

Δ119891 = 25 kHz

(13)


119875

119872

2= 119875

0

2sdot

119903

0

2

119903

410

02120572(119891)119903sdot

119888120591

2

Ψ119903

2sdot 119899120590bs (14)

where 119875



119881 The


Ψ cong

6119888

2

4120587

2119891

2119886

2

(15)

0 200 400 600 8000

100

200

300

400

Frequency (kHz)


120572(d

Bkm

)



119877

119872=

119875

119872

2

119875

119873

2= 3120587

2120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

119886

4119891

4

119903

210

02120572(119891)119903

= 3 times 051

4120587

2119888

4120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

1

Θ

4119903

210

02120572(119891)119903

(16)


120591 = 1ms

119899 = 01mminus3 (119878

119881= 10 log 119904


= minus40 dB)

(17)




0 100 200 300 4000

5

10

15

20

25

30

Range (m)

North

South

Win

Spr

SumAut

Tem

pera

ture

(∘C)

North998400



119860 (119903) = 2int

119903

0

120572 (119903

1015840) 119889119903

1015840 (18)


120572 (119903 119891Area Season)




Δ119860(119903 119891N1015840 Season) = 2int

119903

0

120572 (119903

1015840 119891N1015840 Season) 119889119903

1015840

minus 2119903120572 (50m 119891N Spr)(20)




Δ119860 (119903 119903

119898 119891N Spr) = 2int

119903

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840minus 2119903

sdot

1

119903

119898

int

119903119898

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840

(21)


119898 Although









R (d

B)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60




7

4

8

16

32

14

28






Max

imum

rang

e (m

)

SNR (dB)10

20

30

Frequency (kHz)103

103

102

102

101101

120572 error







SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)









10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)






minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in



119886 cong

051119888

Θ119891

(11)


119877

119865= 4 times 051

6120587

4119888

3120588 sdot

120578119908119863

4120590bs

119873

1198810Δ119891

sdot

1

Θ

6119903

410

02120572(119891)119903

(12)




119888 = 1500ms

120588 = 1025 kgm3

120578 = 06 [3]

119908 = 25Wcm2 [3]

119863 = 1

TS = minus40 dB

119873119881

0= 155 dB re 1 120583Pa2Hz

Δ119891 = 25 kHz

(13)


119875

119872

2= 119875

0

2sdot

119903

0

2

119903

410

02120572(119891)119903sdot

119888120591

2

Ψ119903

2sdot 119899120590bs (14)

where 119875



119881 The


Ψ cong

6119888

2

4120587

2119891

2119886

2

(15)

0 200 400 600 8000

100

200

300

400

Frequency (kHz)


120572(d

Bkm

)



119877

119872=

119875

119872

2

119875

119873

2= 3120587

2120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

119886

4119891

4

119903

210

02120572(119891)119903

= 3 times 051

4120587

2119888

4120588 sdot

120578119908120591119899120590bs119873

1198810Δ119891

sdot

1

Θ

4119903

210

02120572(119891)119903

(16)


120591 = 1ms

119899 = 01mminus3 (119878

119881= 10 log 119904


= minus40 dB)

(17)




0 100 200 300 4000

5

10

15

20

25

30

Range (m)

North

South

Win

Spr

SumAut

Tem

pera

ture

(∘C)

North998400



119860 (119903) = 2int

119903

0

120572 (119903

1015840) 119889119903

1015840 (18)


120572 (119903 119891Area Season)




Δ119860(119903 119891N1015840 Season) = 2int

119903

0

120572 (119903

1015840 119891N1015840 Season) 119889119903

1015840

minus 2119903120572 (50m 119891N Spr)(20)




Δ119860 (119903 119903

119898 119891N Spr) = 2int

119903

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840minus 2119903

sdot

1

119903

119898

int

119903119898

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840

(21)


119898 Although









R (d

B)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60




7

4

8

16

32

14

28






Max

imum

rang

e (m

)

SNR (dB)10

20

30

Frequency (kHz)103

103

102

102

101101

120572 error







SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)









10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)






minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


0 100 200 300 4000

5

10

15

20

25

30

Range (m)

North

South

Win

Spr

SumAut

Tem

pera

ture

(∘C)

North998400



119860 (119903) = 2int

119903

0

120572 (119903

1015840) 119889119903

1015840 (18)


120572 (119903 119891Area Season)




Δ119860(119903 119891N1015840 Season) = 2int

119903

0

120572 (119903

1015840 119891N1015840 Season) 119889119903

1015840

minus 2119903120572 (50m 119891N Spr)(20)




Δ119860 (119903 119903

119898 119891N Spr) = 2int

119903

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840minus 2119903

sdot

1

119903

119898

int

119903119898

0

120572 (119903

1015840 119891N Spr) 119889119903

1015840

(21)


119898 Although









R (d

B)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60




7

4

8

16

32

14

28






Max

imum

rang

e (m

)

SNR (dB)10

20

30

Frequency (kHz)103

103

102

102

101101

120572 error







SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)









10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)






minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


R (d

B)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60




7

4

8

16

32

14

28






Max

imum

rang

e (m

)

SNR (dB)10

20

30

Frequency (kHz)103

103

102

102

101101

120572 error







SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)









10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)






minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in






SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

SNR

(dB)

80

60

40

20

0

minus20

minus40

minus60

TD radius (cm)35

7

4

8

16

3214

28

Beam width (∘)









10 log(

3

25

) + 10 log(

075

06

) + (155 minus 142)

= 148 dB(22)






minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


minus60

minus40

minus20

0

20

40

60

80

SNR

(dB)

Vessel

All

NV0 = 142dB

155dB



0=


0 100 200 300 4000

10

20

30

40

50

60

Range (m)

Abso

rptio

n at

tenu

atio

n (d

B)

WinterSpring

SummerAutumn

200 kHz South

200 kHz North

38kHz North

38kHz South







119873 = 119873

119860+

119873

119879

120578

+ 119873

119867 (119865 minus 1) (23)




1198670119891




119881(1)

in place of 119873119860

119877

119865= 4 times 051

6120587

4119888

3120588

sdot

120578119908119863

4120590bs

Δ119891 [119873

1198810+ 119873

1198790119891

4120578 + 119873

1198670119891

4]

sdot

1

Θ

6119903

410

02120572(119891)119903

(24)







119864119860

119862in (4) large)



0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


0 200 400minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

minus2

minus1

0

1

2

ΔA

(dB)

Range (m)0 200 400

Range (m)0 200 400

Range (m)

0 200 400Range (m)

0 200 400Range (m)

0 200 400Range (m)

Win Spr

Sum Aut

38kHz 70kHz

200 kHz

120 kHz

400 kHz 800 kHz







0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

0 100 200Range (m)

50

100

150

200

70kHz

400 kHz200 kHz

120 kHz

minus2

minus1

0

1ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

minus2

minus1

0

1

ΔA

rm (m)


119898 The






119898 at








4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




4 Conclusions









Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


























Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in

research article effects of noise and absorption on high...

Documents