CCAMLR Science, Vol. 5 (1998): 259-271

SHORT NOTE

ATTRIBUTION OF ACOUSTIC ECHOES TO SQUID IN THE SOUTH ATLANTIC

C. Goss, P. Rodhouse, J.L. Watkins and A.S. BrierleyBritish Antarctic Survey

Natural Environment Research CouncilHigh Cross, Madingley Road

Cambridge CB3 0ET, United Kingdom

Abstract

During a research cruise of the British Antarctic Survey ship RRS James Clark Ross inFebruary 1996, more than 2 000 individuals of the ommastrephid squid, Martialiahyadesi, were caught in a single haul made by a pelagic trawl near the Antarctic PolarFrontal Zone in CCAMLR Subarea 48.3. This paper describes the analysis of acousticdata gathered during this haul and a survey of the immediate vicinity, in order tocharacterise an acoustic signature that may be used to identify this species of squid.Distinctive echoes that did not occur when other, unsuccessful hauls were undertaken,were the only strong echoes recorded from the part of the water column traversed bythe net. The trawl was restricted to a single depth zone which reduced the potentialnumber of echoes that might otherwise have been attributed to the squid that werecaught.

Plus de 2 000 calmars ommastrephid~s, Martialia hyadesi, ont ~t~ captures darts un chalutp~lagique pros de la zone frontale polaire antarctique dans la sous-zone 48.3 de laCCAMLR au cours d’une campagne de recherche qui a ~t~ men~e en f~vrier 1996 par lenavire de recherche du British Antarctic Survey, James Clark Ross. Cette communicationa pour but de d~crire l’analyse des donn~es acoustiques qui ont ~t~ relev~es pendant lechalutage et lors d’une campagne de recherche men~e darts le secteur proche de cettezone en vue de d~finir le caract~re acoustique de cette esp~ce de calmar afin de pouvoir~ventuellement l’utiliser pour l’identifier. Les seuls ~chos intenses qui ont ~t~enregistr~s dans la partie de la colonne d’eau qui traverse le filet sont des ~chosdistinctifs qui ne se sont pas reproduits lors d’autres chalutages dans lesquels la capturede cette esp~ce a ~t~ nulle. L’op~ration de chalutage n’a ~t~ r~alis~e que dans une seulezone de profondeur, ce qui a r~duit le nombre d’~chos potentiels qui auraient autrementpu ~tre attribu~s aux calmars captures.

Pe3roMe

B xo~e nayqHo-nccne)!oBaTeabcKoro pefica 6pnTartcKoro cy;~Ha James Clark Ross,npoBe;~ennoro B ~eBpaae 1996 r. B AH~apKTnqeCKO~ noaapHO~ ~ponwaa~no~ 3OHeHoApafiona AHTKOMa 48.3, 3a OAHO ~paaenne pa3Horay6nnn~iM ~paao~ 6blaOBB~aOBaeHO CBbIme 2000 KaabMapoB Bn~a Martialia hyadesi. B ~aHHO~ pa6o~enpeAcwaBfleHBi aKycwnqecKue Aann~ie, no~yqeHHbie B xoae a~oro ~pafleHna n c~eMKn,BbmO~eHno~ B 6an3ae~amnx Borax, C ~e~b~ onpeae~ aKyCT~qecK~enapaMeTpbi, KOTOpbIe MO~HO 6yae~ RCHOflb3OBaTb ~fla n~euT~nKaRnn ~aHHOFOB~a Ka~bMapa. qewK0 pa3~nqam~nec~ 3XOCHFHa~BI, KOTOpblX He Ha6~m~a~ocb BTpa~eHnaX 6e3 BbI~OBa H~H C He60~bmnM BbI~OBOM Ka~bMapa, ~BH~HCb

e~HHCTBeHHblMH CH~bHblMH c.rna~aM~, 3aperncTp~poBannbiMn B To~eo6flaB~BaeMofi Tpa~oM BO~bI. Tpaa pa6oTaa TO~bKO Ha O~HOM rop~3OHTe, qTOCoKpaT~O BO3MO~HOe KO~HqeCTBO TeX CRFHa~OB, KOTOpble B HpOTHBHOM c~yqaeMO~HO ~bl~O ~bI np~nnCaTb BbI~OB~eHHblM Ka~bMapaM.

259

Goss et al.

Resumen

Durante una campafia de investigaci6n britfinica efectuada con el RRS James Clark Rossen febrero de 1996 en la Subirea 48.3 de la CCRVMA, mils de 2 000 espedmenes delcalamar omastr~fido Martialia hyadesi fueron capturados en un arrastre realizado enaguas peligicas cerca de la Zona del Frente Polar. Este trabajo describe el anilisis de losdatos acfsticos obtenidos durante este arrastre y durante una prospeccidn en aguasadyacentes a fin de determinar el perfil acfstico caracterfstico que se puede utilizar paraidentificar esta especie de calamar. Los ecos caracteristicos, que no ocurrieron cuando serealizaron otros lances sin ~xito, fueron las tinicas sefiales intensas registradas en lacolumna de agua recorrida por la red. E1 arrastre se limit6 a un estrato de profundidad,de modo que se redujo el mimero potencial de ecos, que de otra manera podrian habersido atribuidos al calamar que fue capturado.

Keywords: squid, ommastrephids, hydroacoustics, pelagic trawl, ehco attribution,target strength, South Atlantic, CCAMLR

INTRODUCTION

Analysis of the diet of albatrosses and elephantseals breeding on the South Atlantic Island ofSouth Georgia, led scientists at the BritishAntarctic Survey (BAS) to realise that there mustbe substantial stocks of squid available within theforaging range of these predators (Rodhouse etal., 1993; Rodhouse et al., 1996). The squid takenmost often by predators were not the twocommonly fished species in the South Atlantic(Illex argentinus and Loligo gahi), but a thirdspecies, sometimes caught as a by-catch with Illexargentinus, the circumpolar ommastrephidMartialia hyadesi. Apart from data collected in1989, when large numbers of M. hyadesi werecaught by a commercial jigger during exploratoryfishing trials in the Antarctic Polar Frontal Zone(APFZ) (Rodhouse, 1991), our knowledge of thedistribution and life history of this species in theCCAMLR area is based only on data from catchesof small numbers of individuals (Rodhouse et al.,1996), and recent data from a new fisherydeveloping at South Georgia (Gonzfilez andRodhouse, in press; Rodhouse, 1997). Excludingthe commercial fishery on the Patagonian Shelf(see Gonzfilez et al., 1997), which is beyond theforaging range of the major Antarctic predators(other than the wandering albatross), M. hyadesihas been difficult to sample and, as aconsequence, there is a lack of information aboutthe acoustic characteristics of this species. Agrowing amount of information has beencollected about the location and timing of thefeeding of grey-headed albatrosses (a speciesknown to depend heavily on M. hyadesi) usingsatellite tags; this has pointed to the APFZ as themost probable site from which birds obtainM. hyadesi (Rodhouse et al., 1996).

In 1996 more than 2 000 individuals ofM. hyadesi were caught in a pelagic trawl near theAPFZ during a research cruise of the BAS shipRRS James Clark Ross. This paper examinesacoustic data gathered during this haul in anattempt to describe echochart marks that can beattributed to squid. Although the haul covered anextensive horizontal track, it was carried out at asingle depth, which reduced the number ofpotential acoustic targets that could be attributedto the haul. No other species were caught withthe squid which is an advantage for making apositive identification of echoes. An initial reviewof the echocharts showed fairly intense, scatteredmarks that are unlike the shallow aggregations,attributed to krill, usually seen in the vicinity ofSouth Georgia. The absence of any other strongechoes during the haul at the depth fishedincreases the likelihood that these echoes arecaused by squid. In other hauls in the area, alsodescribed here, only small numbers of squid werecaught; these did not coincide with any strongacoustic echoes. If this acoustic characterisationcan be confirmed, it will be possible to useacoustic surveys to further describe thedistribution and behaviour of M. hyadesi stocks(see Starr and Thorne, in press, for a review ofacoustic assessments of squid).

METHODS

In early February 1996 the RRS James Clark Rosstraversed a region of the South Atlantic Ocean inthe vicinity of the Maurice Ewing Bank where theAPFZ had been located a month before (Figure 1).Satellite tags on grey-headed albatrosses in theprevious year had suggested that this area wasbeing actively targeted by foraging birds, and wastherefore a probable site for M. hyadesi (Rodhouseet al., 1996).

260

Attribution of Acoustic Echoes to Squid in the South Atlantic

I

60°W

"%

!50 °W

_ 50°S

Falkland/Malvinas Islands

!40°W

//-20oo~ Maurice/ Ewing Bank

!

~_.. South Georgia.___

I

Figure 1: Map of the South Atlantic Ocean showing the context of the squid study, the Antarctic Polar FrontalZone (APFZ) and study area (stippled).

The aims of the acoustic study were, firstly, tohelp determine the location of suitable targets forsquid fishing and secondly, to characterise theechosignals of any squid caught in a pelagic net.The pelagic trawl, towed at between three andfour knots, was designed for catching small fish,and has been described previously by Rodhouseeta!., 1996. Some modifications had been made tothe net before the present survey, to increase thenet opening to 25-30 m in height. A depth sensorwas attached to the net and this was used toestimate the net depth using an acoustic link (Roeand Shale, 1979).

A Simrad EK500 echosounder (software v.4.01)operating at 38 kHz, 120 kHz and 200 kHz wasused during a series of transects and net haulswithin the survey area (stippled in Figure 1).Echosounder settings are summarised in Table 1.

Table 1: Echosom~der setth~gs.

Sounder Data Simrad EK500

Frequency, kHzPing rate, sPulse length, msBand width, kHzBeam angle, °Threshold dBNoise margin, dBEchochart minimum, dB

38 1204 4

1.0 1.03.8 1.2

7.O2 9.10-100 -100

0 0-75 -75

Integration of acoustic backscatter down to1 000 m was carried out using 5 m depth layersand 100-second horizontal bins (resets). Theecho strength data are expressed in cubic metresas mean volume backscattering strength (MVBS)in dB//1 m for each bin. The sounder wascalibrated using three sizes of copper spheresuspended around 20 m below the ship’s hull inaccordance with a standard method (MacLennanand Simmonds, 1992 and Simrad EK500 operatormanual). Calibrations were carried out in adeep, sheltered inlet at South Georgia on twooccasions in mid-January and mid-February1996, and average values were used for the gainsettings. Sound speed and absorption coefficientsettings were calculated for the conditionsencountered in the waters around South Georgiabut no changes were made during the cruise tosimplify data collection. Integrated data wereprocessed to add these calibration correctionsbut no additional corrections have been made forthe elevated temperatures encountered in part ofthis fishing area because of the complexoceanographic conditions there, although thesewould be necessary if the acoustic data werebeing used to estimate biomass. TVG-amplifiednoise was removed from the integrated resultsusing the visualisation package AVS asdescribed by Socha et al. (1996) and Watkins andBrierley (1996).

261

Goss et al.

RESULTS

During the night of 11/12 February 1996 thepelagic trawl was fished across the area at a depthof 25 m for 12 hours and caught an estimated2 300 M. hyadesi (Table 2). No other species wererecorded in this haul. The average mantle lengthof squid in this catch was 228.6 mm _+ 21.8(standard deviation). The echocharts recorded at38 kHz and 120 kHz during part of this haul arereproduced as Figure 2. Most of the haul (the firstseven hours) was through an area notable for thestrongly speckled layer on the echocharts(subsequently referred to as speckles) between thesurface and 50 m. The charts from the twofrequencies were recorded using the same colourminimum (-75 dB), so that the level of the weakestvisible echoes, and the level indicated by eachcolour range, are the same for both. However,calibration constants from a calibration carriedout in the UK had been used during the surveyperiod, so that it was necessary to makeadjustments of +0.92 dB for 38 kHz and +2.5 dBfor 120 kHz to the raw data. Thus the trueintensity range for the 120 kHz chart starts about1.6 dB below that of the 38 kHz chart. Taking thisinto account, it is evident from the charts that thislayer scattered more weakly at 120 kHz. Afterdawn, the speckles cleared away and no moretargets were seen.

The only other class of marks seen during thetrawl is also illustrated in the sections of chartin Figure 2: slightly more compact, extendingbetween about 40 m and 240 m in depth andof varying density, these loose aggregations areonly marginally stronger on this 38 kHz chartthan on the 120 kHz chart. After adjustment,echo-integrator values from the vicinity of thesemarks are about +1 dB higher at 120 kHz than at38 kHz.

The echoes caused by squid during daylightmay appear very different on the charts from thenear-surface speckled marks of Figure 2. Figure 3shows a chart collected in the fishing area at duskon which strong aggregations, which had beenregularly seen in midwater during daylight hoursin the area, appear to migrate upwards anddissipate, leaving the speckled layer as seen atnight in Figure 2. The loose (+1 dB) aggregationsalso developed on this occasion as the sun set, butit is not clear which of the daylight marks (if any)can be attributed to the same source.

Five other hauls carried out in the study areawith the pelagic trawl that caught only smallsamples of M. hyadesi (less than 40 individuals)(details in Table 2) did not pass through eithernight-time speckled layers or the daytime compactmarks as shown on the charts in Figure 2. Thepresence of some squid in these nets suggests that

Table 2: Haul data.

Net Date Latitude Depth (m) Martialia Other SquidTime Longitude Duration hyadesi

(GMT)

1 11 Feb 96 50°12.85’S 210 - 300 0 011:10 40°47.18’W 02:10

2 11 Feb 96 50°16.98’S 25 2 310 023:00 40°37.27’W 12:10

3 13 Feb 96 50°09.58’S 20 3O 401:51 40°43.51’W 01:30

10002:25

4 13 Feb 96 49°32.03’S 130 28 517:52 39°56.08’W 01:45

8000:3020

01:55

5 14 Feb 96 49°36.35’S 150 7 008:40 39°49.83’W 01:30

6 16 Feb 96 49°41.60’S 60 36 600:30 39°31.86’W 00:45

2002:00 Extensive net

70 - 100 damage02:10 sustained

262

120

kHz

38 k

HZ

00

o o

a.qu

eDv

tI1no

s aq

l tq

p~.n

b S o

i sao

qa~t

a.q

snoa

v j

o uo

!lnq~

.a3,

1 v

100

200

4OO

Sunset

Figure 3: Echochart recorded at 38 kHz in the fishing area at dusk.

Attribution of Acoustic Echoes to Squid in the South Atlantic

the fishing was in the same area as M. hyadesiaggregations, but that major concentrations hadbeen bypassed. Clear targets where TS at 38 kHzwas higher than 120 kHz were visible during thesection of the first transect that preceded the firsthaul. However, during the daytime tow thestrong marks stayed below the 300 m maximumfished by the net for all but a very short portion ofthe haul. The second haul was the successfulnight haul described in the preceding paragraph.During the third, night-time haul, the speckledlayer was replaced by a concentrated +1 dB layer,at around 40 to 50 m water depth. During thelatter part of this haul at 100 m the net probablypassed through acoustic targets represented bydiffuse irregular +1 dB patches. The fourthdaytime haul was towed at 130 m, 80 m and 20 m,but all of these sections of the haul were probablytoo shallow to catch the targets causing the onlystrong marks that appeared on the echocharts,which were between 150 m and 250 m. Similarly,the only strong marks around the fifth night-timetow appeared before deployment and after thenet had been hauled in. During the final haul thenet suffered extensive damage, thus the size ofthe catch cannot be related to the acoustics, butthe catch of M. hyadesi provides additionalconfirmation of its presence in the area.

Integrated acoustic results were used to derivethe quantity 3 MVBS: the number of decibelsremaining when MVBS at 38 kHz is subtractedfrom MVBS at 120 kHz. Figure 4a shows thefrequency distribution of this quantity for all theintegrator observations between 20 m and 50 m forthe main part of the haul that contained targets.Figure 4b shows the frequency distribution for asubset of these observations; this subset waschosen by referring to the echocharts, andselecting a portion of the tow when speckles werethe only evident mark.

Observations During Transects

The first long transect during the survey beganwith similar acoustic layers to earlier transectsmade to the north of the APFZ: a deep layervisible on 38 kHz charts, extending through muchof the area, at depths of 500 to 600 m. The widthof the sounder beam at this depth means that littlehorizontal or vertical detail can be detected. Atnight, north of the APFZ, the shallow, speckledlayer, characteristically stronger on 38 kHz than120 kHz, and seen during the major squid haul,extended down to 50 m on a number of transects.During the daytime, diffuse +1 dB patches below

50 m were common, having a characteristicallyeven intensity. Interspersed patchily with thesemarks were small, dense, irregular aggregationswith a similar frequency difference to thenight-time, near-surface speckles.

Squid Target Strength (TS)

A summary of previously published dataon squid TS is given in Table 3. Because squiddo not have a swim bladder, their TS is lowin comparison with fish of equivalent sizepossessing swim bladders. Arnaya et al. (1988,1989a) distinguished between intensity TS andenergy TS, the former representing peak values,and the latter, integrated values of the wholeecho. They recommend the use of energy TSbecause this will take into account variations withpulse length and wave length; this type ofmeasurement is thought to be more accurate thanthe intensity TS. These authors present separateregression equations for the two types of TSversus mantle length at two frequencies, but theyalso state that a linear relationship between thetwo types of measurement could not beestablished. Their intensity TS values have beencited in Table 3 to allow comparison with earlierpeak-based measurements. A number of TSestimates of various squid species have shown theimportance of using live, free-swimmingspecimens for this purpose; Arnaya et al. (1989b)recorded much lower values with live squid thanthey had using dead specimens in their earlierstudies, making their results comparable withthose recorded by Jefferts et al. (1987) for Loligoopalescens. Arnaya et al. (1988, 1989a, 1989b, 1989c)and Arnaya and Sano (1990a, 1990b) have studiedthe effect of aspect and swimming speed on TS,and in the last paper in the series compare theirlatest results with predictions from a scatteringmodel, achieving good agreement between theirobservations and predictions. Because of thesimilar size (230 mm) of the squid and the factthat live, encaged squid were used, the results ofArnaya et al. (1989c) are considered the mostappropriate for the squid caught in this survey.Until TS measurements of M. hyadesi becomeavailable, this would suggest a TS around -46 dBat 38 kHz.

DISCUSSION

Speckled marks that dominated theechotraces during the haul in which largenumbers of squid were caught are thought to becaused by M. hyadesi. The ~ MVBS (MVBS at

265

Freq

uenc

yFr

eque

ncy

o0

Attribution of Acoustic Echoes to Squid in the South Atlantic

Table 3: Squid target strength from published sources.

Species Lengths Frequency TS in dB Source(dorsal mantle (kHz) (max. and rain.length in cm) where two values

are given)

Loligo opalescens 4.5 - 16.0 200 -49.3 -38.3 Vaughan (1978)(frozen and thawed)

Loligo opalescens 11.6 50 -58.6 -58.7 Jefferts et al. (1987)Todarodes pacificus 14.8 - 29.2 50 -46.3 -32.3* Arnaya et al. (1988)Ommastrephes bartrami 15.2 - 28.1 20O -42.4 -30.0* Arnaya et al. (1989a)(frozen and thawed)

Todarodes pacificus (live) 16.0 28.5 -51.35 Arynaya et al. (1989b)96.2 -54.60

Todarodes pacificus 23.67 28.5 -45.66 Arnaya et al. (1989c)50 -46.5396.2 -48.04200 -47.62

Ommastrephes bartrarni 20* 28.5 -41" Kajiwara et al. (1990)Loligo edulis 4-16 20O -53.20 -40.70 Lee et al. (1991)(dorsal aspect)

Loligo edulis 4-9 420 -69.5 -57.5 Lee et al. (1992)(in situ, dual-beam) 10 - 15 -55.5 -45.5

* Inferred from published regression equation

120 kHz - MVBS at 38 kHz) values measuredwhen these marks predominate is largely between-3 dB and +1 dB, which leads us to conclude thatthe layer is composed of scatterers larger thankrill (see Madureira et al., 1993, for a descriptionof the frequency difference in backscatteringstrength among different classes of scatterers).The +1 dB 3 MVBS values predominate in theoverall frequency distribution shown in Figure 4a,but there is a shoulder on the negative side of thepeak of the distribution, which is reflected as aseparate peak (3 MVBS -- -1) in the speckle-onlysubset (Figure 4b). These observations areconsistent with the targets being composed of amixture of two size classes of organism, forexample squid and small fish.

The description of night-time echoes thought tobe caused by M. hyadesi in this paper correspondsvery well with the appearance of night-timeechoes, on charts produced with similarequipment (Simrad EY500, operating at 38 kHz) insurveys off the coast of Peru, that have beenattributed to Dosidicus gigas (Ommastrephidae)(Torero, Instituto del Mar del Peru, pers. comm.).

The overall levels of MVBS throughout fishingwere low, only rarely exceeding -70 dB at 38 kHz.If the TS of squid is around -46 dB, this wouldsuggest a very low density indeed, but the extentof the marks and the duration of the haul makesthe attribution of the marks feasible for the size of

catch. In spite of the long duration of the haul inwhich M. hyadesi were caught, no other markswere visible; this is also taken as confirmation ofthis identification.

The +1 dB aggregations are not thought to becaused by squid because these marks only lastedfor the first two or three hours of the major haul,and many of the individuals in the catch were invery good condition, still pulsating when broughton deck - which is not thought to be likely if theyhad been in the trawl for 10 hours. A possiblesource of the marks may be the myctophid prey ofthe squid, also caught in the area on this surveyusing a 25 m2, rectangular midwater trawl. Thedecibel difference for this class is marginally lessthan the +2 dB minimum used to identify krill(Madureira et al., 1993). Other marks recordedduring the search for squid had a positivedifference in the integrated backscattering levelsat 120 kHz and 38 kHz; these are thought to havebeen produced by Euphausia frigida (cf. Brierley etal., in press).

Assuming that squid in this area are the sourceof the speckled layer at night, and are moreaggregated during the day, as suggestedby the echoes in Figure 3, there are three majoradvantages of detection at night. The firstis that distributions recorded at night areshallow, which is a fundamental advantage forhull-mounted or near-surface transducers because

267

Goss et al.

interference from background noise, amplified bythe time-varied gain of the sounder, is at a lowerlevel than at greater depths. Near-transducerdistributions are also more precisely mapped thanthose at a distance from the transducer where thebeam has spread, increasing the volume of thebasic echo sample. The second is that diffuselayers are more widespread than compactaggregations, and are therefore more frequentlyencountered for a given survey distance. Thethird useful feature of the echoes seen here is theirdistinctiveness; compact aggregations seen duringthe daytime would need to be distinguishedfrom krill and possibly fish. Searching andcharacterisation of squid echoes is thereforeexpected to be most effective at night. Oncedetection and identification have been achieved,acoustic stock assessment might be carried outduring the daytime so that no individuals areexcluded because they are shallower than thetransducers (at least in standard configurations)or because they are too diffuse to be detected.

CONCLUSIONS

A single large haul of M. hyadesi was caughtduring the survey at night; a number of featuresof the haul enabled echoes observed while it wasunder way to be attributed to the squid with agood degree of certainty. These are that:

¯ no other species were caught with the squid;

the haul was made at a single, shallow depth.Many echoes appearing on the charts in thisarea were confined to a particular depthhorizon, and by fishing at a single depth thenumber of potential candidate echo-types wasreduced to two;

one of the potential echo-types that may havebeen caused by squid was only recorded at thestart of the very long haul. Squid caught atthat time would have been in very poorcondition when they were landed, but thesquid actually caught were very fresh;

the difference in echo strength at two acousticfrequencies confirmed that echoes causing themost likely echo trace also had acousticcharacteristics appropriate for this size oftarget organism; and

a number of additional small catchesof M. hyadesi made during the survey weremade in places and at depths where thecharacteristic echoes were not present.

The characterised echo traces were notobserved during the daytime, although compact,deeper marks not seen at night were observed.Further fishing to identify daytime echo tracesand to confirm the attribution of echoes recordedat night would be very valuable.

A review of the published target strengthsestimated for squid demonstrated the variabilityof this quantity, and the importance of makingnew estimates, especially for M. hyadesi. Using apublished target strength value for squid of asimilar size to those caught in this surveysuggested that the squid caught had been in awell dispersed layer, but that the extent of thelayer was sufficient to produce the size of catchactually observed.

ACKNOWLEDGEMENTS

We acknowledge the assistance of Peter Boyle,Heather Daly and of the master, officers and crewof RRS James Clark Ross. The comments from tworeferees were also helpful.

REFERENCES

Arnaya, N. and N. Sano. 1990a. Studies onacoustic target strength of squid, V. Effect ofswimming on target strength of squid. Bull.Fac. Fish., Hokkaido Univ., 41 (1): 18-31.

Arnaya, N. and N. Sano. 1990b. Studies onacoustic target strength of squid, VI.Simulation of squid target strength by prolatespheroidal model. Bull. Fac. Fish., HokkaidoUniv., 41 (1): 32-42.

Arnaya, N., N. Sano and K. Iida. 1988. Studies onacoustic target strength of squid, I. Intensityand energy target strengths. Bull. Fac. Fish.,Hokkaido Univ., 39 (3): 187-200.

Arnaya, N., N. Sano and K. Iida. 1989a. Studieson acoustic target strength of squid, II. Effectof behaviour on averaged dorsal aspect targetstrength. Bull. Fac. Fish., Hokkaido Univ.,40 (2): 83-99.

Arnaya, N., N. Sano and K. Iida. 1989b. Studieson acoustic target strength of squid, III.Measurement of the mean target strength ofsmall live squid. Bull. Fac. Fish., HokkaidoUniv., 40 (2): 100-115.

268

Attribution of Acoustic Echoes to Squid in the South Atlantic

Arnaya, N., N. Sano and K. Iida. 1989c. Studieson acoustic target strength of squid, IV.Measurement of the mean target strength ofrelatively large-sized live squid. Bull. Fac.Fish., Hokkaido Univ., 40 (3): 169-181.

Brierley, A.S., P. Ward, J.L. Watkins and C. Goss.In press. Acoustic discrimination of SouthernOcean zooplankton. Topical Studies inOceanography, Deep-Sea Research, Part II.

Gonz~lez, A.F. and P.G. Rodhouse. In press.Fishery biology of the sevenstar flying squidMartialia hyadesi at South Georgia duringwinter. Polar Biol.

Gonz~lez, A.F., P.N. Trathan, C. Yau andP.G. Rodhouse. 1997. Interactions betweenoceanography, ecology and fishery biology ofthe ommastrephid squid Martialia hyadesi inthe South Atlantic. Mar. Ecol. Prog. Ser., 152:205-215.

Jefferts, K., J. Burczynski and W.G. Pearcy. 1987.Acoustical assessment of squid (Loligoopalescens) off the central Oregon coast. Can. J.Fish. Aquat. Sci., 44: 1261-1267.

Kajiwara, Y., K. Iida and Y. Kamei. 1990.Measurement of target strength for the flyingsquid (O~nmastrephes bartrami). Bull. Fac. Fish.,Hokkaido Univ., 41 (4): 205-212.

Lee, K.T., W.H. Shih, C.H. Liao and S.S. Chyn.1992. Application of dual-beam acousticsurvey techniques to assess the sizedistribution of squid, Loligo edulis. Journal ofthe Fisheries Society of Taiwan, 19: 25-34.

Lee, K.T., W.H. Shih, C.H. Liao and J.P. Wang.1991. Studies on the hydroacoustic scatteringproperties of individual squid, Loligo edulis.Journal of the Fisheries Society of Taiwan, 18:215-225.

MacLennan, D.N. and E.J. Simmonds. 1992.Fisheries Acoustics. Chapman and Hall, London.

Madureira, L.S.P., I. Everson and E.J. Murphy.1993. Interpretation of acoustic data at twofrequencies to discriminate between Antarctickrill and other scatterers. J. Plankton. Res., 15:787-802.

Rodhouse, P.G. 1991. Population structure ofMartialia hyadesi (Cephalopoda, Omma-

strephidae) at the Antarctic Polar Front andthe Patagonian Shelf, South Atlantic. Bulletinof Marine Science, 49: 404-418.

Rodhouse, P.G. 1997. Precautionary measures fora new Martialia hyadesi (Cephalopoda,Ommastrephidae) fishery in the Scotia Sea: anecological approach. CCAMLR Science, 4:125-139.

Rodhouse, P.G., J.P. Croxall and P.A. Prince.1993. Towards an assessment of the stock ofthe Ommastrephid squid Martialia hyadesi inthe Scotia Sea: Data from predators. In:Okutani, T., R.K. O’Dor and T. Kubodera(Eds). Recent Advances in Cephalopod FisheriesBiology. Tokai University Press, Tokyo:433-440.

Rodhouse, P.G., P.A. Prince, P.N. Trathan,E.M.C. Hatfield, J.L. Watkins, D.G. Bone,E.J. Murphy and M.G. White. 1996. Cepha-lopods and mesoscale oceanography at theAntarctic Polar Front: satellite-trackedpredators locate pelagic trophic interactions.Mar. Ecol. Prog. Set., 136: 37-50.

Roe, H.S.J. and D.M. Shale. 1979. A new multiplerectangular midwater trawl (RMTI+8M) andsome modifications to the Institute ofOceanographic Sciences RMTI+8. Mar. Biol.,50: 283-288.

Socha, D.G., J.L. Watkins and A.S. Brierley. 1996.A visualization-based post-processing systemfor analysis of acoustic data. ICES J. Mar. Sci.,53: 335-338.

Starr, R.M. and R.E. Thorne. In press. AcousticAssessment of Squid Stocks. In: Rodhouse,P.G., E.G. Dawe and R.K. O’Dor (Eds). SquidRecruitment Dynamics, FAO Fisheries TechnicalPaper.

Vaughan, D.L. 1978. The target strength ofindividual market squid Loligo opalescens.Calif. Dept Fish and Game, Fish Bull., 169:149-159.

Watkins, J.L. and A.S. Brierley. 1996. A post-processing technique to remove backgroundnoise from echo integration data. ICES J. Mar.Sci., 53: 339-344.

269

Goss et al.

Tableau 1:

Tableau 2:

Tableau 3:

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Ta6~ima 1 :

Ta6.~uua 2:

Ta6.rI.tla 3:

PFICyHOK 1 :

PI~CyHOK 2:

PHCyHOK 3:

PHCyHOK 4:

Tabla 1:

Tabla 2:

Tabla 3:

Figura 1:

Liste des tableaux

R~glage de l’6cho-sondeur.

Donn~es provenant des chalutages.

Intensit~ de la r~ponse acoustique des calmars calcul6e a partir des informations publi6es.

Liste des figures

Carte de l’oc~an Atlantique sud indiquant le contexte de l’6tude r6alis~e sur les calmars, la zonefrontale polaire antarctique et le secteur d’6tude (hachure).

Enregistrements acoustiques effectu~s h 38 kHz et 120kHz pendant une pattie du chalutagep61agique au cours duquel la majorit~ de M. hyadesi a ~t~ captur~e.

Enregistrements acoustiques h 38 kHz dans le secteur de p~che au cr~puscule.

(a) Distribution de fr~quences de l’intensit~ moyenne de r~trodiffusion par volume 5 pendant lacapture la plus importante de M. hyadesi.

(b) Distribution de fr~quences de l’intensit~ moyenne de r~trodiffusion par volume 5 dusous-~chantillonnage h taches.

~eTa.rlg tlacTpOgIKl4 3xo.rIOTa.

~aHttbie nO Tpa.rleHU~IM.

C~33a ~Ie33b~ KanbMapa - ~3 ony63mKOBa..b~x .CTOq~I~KOB.

CHIICOK pUCyHKOB

KapTa m~HO~ uacxu AT~Ja~tT~qeCKOrO oKeaHa, noKa~bma~oma~ AHTap~uqecKy~o nonnp~y~oqbpo~xanbHy~O ~OHy (APFZ) u pa~oH nccne~o~aHnn (TOqKn).

:)xorpaM~bI, nony’JeHHb~e Ha 38 ~ 120 KF~I S xoAe xo~ qaCTU Tpa~eHU~, KOrZla 6b~a nonyqena66nbman qaCTb BbI~OBa M. hyadesi.

:3xorpaMMa, no~yqeu~bm ~a 38 KF~I s pa~one npoMb~cna ~ cyMepKax.

(a) HaC~OTHOe pacnpez~e:~e~ue 5 MVBS go ~peMs Tpa~euun, ~asmero Han6o~ee KpynHbI~Bb~,qO~ M. hyadesi.

(b) ~IaCTOTnOe pacnpe;leneHue 5 MVBS BO S~TO~ SbI6OpKe ~xoc~r~anoB.

Lista de las tablas

Ajustes de la ecosonda.

Datos del lance.

Fuerza del blanco del calamar obtenido de la literatura.

Lista de las figuras

Mapa del Oc6ano Atlfintico sur mostrando la ubicaci6n del gtrea de estudio del calamar, la Zona delFrente Polar (APFZ) y el ~irea de estudio (punteada).

270

Figura 2:

Figura 3:

Figura 4:

Attribution of Acoustic Echoes to Squid in the South Atlantic

Diagramas del eco registrado a 38 kHz y 120 kHz durante parte del arrastre pelfigico en que secaptur6 la mayor parte de M. hyadesi.

Diagrama del eco registrado a 38 kHz en la zona de pesca, al atardecer.

(a) Frecuencia de distribuci6n de 8 MVBS durante la captura mayor de M. hyadesi.(b) Frecuencia de distribuci6n de ~ MVBS de un subconjunto de sefiales.

271