© 2005 International Council for the Exploration of the Sea
The target strength of Antarctic krill (Euphausia superba) measured by the split-beam method in a small tank at 70 kHz
Tokyo University of Marine Science and Technology 4-5-7, Konan, Minato, Tokyo 108-8477, Japan
*Correspondence to K. Amakasu: tel: +81 3 5463 0481; fax: +81 3 5463 0518. e-mail: soundproc{at}yahoo.co.jp.
The target strengths (TS) of Antarctic krill (Euphausia superba) were measured at 70 kHz aboard the research and training vessel "Umitaka-maru" of Tokyo University of Marine Science and Technology in February 2003 during a Southern Ocean survey. The systematic variations of TS vs. the incident angle of the ensonified wave, henceforth called TS-patterns, were successfully measured for 12 live Antarctic krill. These measurements were compared with the theoretical TS-patterns predicted by the Distorted-Wave Born Approximation-based deformed-cylinder model (DWBA model). While there was good agreement near the main lobe, the measurements were higher than the model predictions in the side-lobe regions; this is consistent with the observations of others. Several possible causes of this discrepancy such as the bending of abdomen and scattering from pleopods were examined, but no single factor was identified as the cause. Rather, it is likely that the discrepancy is a result of a combination of several factors.
Keywords: Antarctic krill, small tank, split-beam method, target strength
Received 24 May 2004; accepted 12 July 2005.
 |
Introduction |
|---|
 Top Introduction MethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
Antarctic krill (Euphausia superba) is a key species in the Antarctic ecosystem and it is also important as one of the region's fisheries resources. Therefore, acoustic surveys using quantitative echosounders have been conducted to evaluate the Antarctic krill stocks (Hewitt et al., 2002).
The target strength (TS) is very important as the scaling factor to convert to "abundance" the acoustic outputs obtained by the quantitative echosounder. The TS of Antarctic krill, however, varies according to many factors, such as orientation, length, density, and sound-speed in the body. The orientation dependence of TS (TS-pattern) is especially important and needs to be clarified, because TS must be averaged over the orientation distribution to obtain the actual scaling factor. Also, it is necessary to clarify the TS-pattern by measurement to verify the theoretical models that are used to estimate the TS under various conditions. However, measurements of the TS-patterns for Antarctic krill are difficult, because the TS per se is variable and small as a result of the small, fragile, and complex body shapes without bony structures and swimbladders. Additionally, it is difficult to get live samples and to control orientation by fishing lines as is done for TS measurements of fish (Sawada et al., 1999).
Although, the TS of Antarctic krill at 120 kHz has been measured already (Foote et al., 1990; Hewitt and Demer, 1991; Pauly and Penrose, 1998), the TS-pattern itself was not obtained experimentally. McGehee et al. (1998) conducted TS-pattern measurements for Antarctic krill, which were tied on the beam axis with a monofilament line, using a single-beam transducer at 120 kHz. Their results showed that the measured TS agreed with the main lobe of the Distorted-Wave Born Approximation-based deformed-cylinder model (DWBA model) predictions. The measured TS, however, was higher than the model predictions in the side-lobe regions. The authors reported that this discrepancy could not be attributed entirely to noise because the measurements were higher than the noise threshold. The slight difference of shape and material property between the model and the real animal was mentioned as the cause of discrepancy.
Demer and Conti (2003) developed the stochastic version of the DWBA model (SDWBA model) which takes account of the phase variability in the solution of the DWBA model; the SDWBA model showed reasonable agreement with the measured TS of McGehee et al. (1998) in the side-lobe regions. They proposed three physical reasons for the apparent variability in the phases of scatter from scattering elements of the krill: first, scattering in a field with noise is a stochastic process; second, krill have shapes that are more complex than juxtaposed cylinders of varying radii; and third, their bodies flex as they swim. These and other potential sources of phase variability, however, have not been individually examined.
We developed a new, precise, and convenient method for TS-pattern measurements of small-sized animals such as Antarctic krill. The orientation-measurement approach of McGehee et al. (1998) was extended using the split-beam method. We confirmed the method as able to measure the TS-pattern accurately by preliminary TS measurements using a physical model and shrimps in our previous work (Amakasu et al., 2003). The TS measurements using this system can be conducted on board a ship where it is easy to get live samples.
The purposes of this study are to measure the TS-pattern of live Antarctic krill precisely at 70 kHz, a candidate frequency for krill surveys, by the new method on board a research vessel in the Antarctic, and also to examine the causes of TS variability from various physical viewpoints.
|
Methods |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
TS- and orientation-measurement systems
The measurement system is shown in Figure 1 (Amakasu et al., 2003). It comprises a 70-kHz, split-beam echosounder (TS-measurement system) and a video camera (orientation-measurement system) to measure the TS and the orientation of animals simultaneously. The transducer was set facing downwards at the top of a small tank (1 x 1 x 1 m) with a window on one of the side walls for observing the orientation. Since this system could be set up indoors in a small space such as on board a ship, external disturbances by wind, waves, and other animals were avoided. The two measurement systems were synchronized to combine the TS and orientation data in post-processing.
|
The specifications of the 70-kHz split-beam echosounder are shown in Table 1. A frequency of 70 KHz was selected because the TS variations caused by the variation in animal orientation may be smaller at 70 kHz than at 120 kHz, the frequency often used in krill surveys (Furusawa et al., 1994). The split-beam method allowed us to measure TS precisely, and a target did not need to be on the beam axis and was only tethered lightly by a thin (e.g. 0.14 mm diameter) monofilament line to make it stay in the beam (Figure 1). Therefore, the animal was able to swim almost freely, hover, or change orientation during the measurements. Further, the orientation angle could be measured in almost all cases, because the split-beam method was employed not only for transducer-directivity correction but also for orientation measurement. The measurement system was calibrated with a 38.1-mm tungsten-carbide sphere before the TS experiments (Foote et al., 1987; Furusawa et al., 1995). The reflection from the tank walls and the reverberation were eliminated by the coherent subtraction technique (Ding, 1997). The range of the target from the transducer was greater than 60 cm and the far-field condition was confirmed.
|
The target orientations were recorded during the TS experiments by video camera (Sony Handycam, CDR-TRV7, 680 000 pixels per frame), which was set close to the observation window (Figure 1). The tilt angle was calculated in post-processing by extending the orientation-measurement method of McGehee et al. (1998) via information from the split-beam method. Hence, the tilt angle,
t, at any position in the beam was calculated by
|
| (1) |
is the orientation vector of Antarctic krill, running from the end of the cephalothorax to the end of the eye, as shown in Figure 2, 
is the wavenumber vector to the target animal. Positive t denotes "head up" and a negative value "head down". The vectors and were calculated from the position data obtained by the split-beam method and the video image. The TS-patterns were obtained by making the TS a function of t in the post-processing.
|
Theoretical model
The DWBA model (Chu et al., 1993; Stanton et al., 1993, 1998; McGehee et al., 1998) was compared with the measured TS to confirm reliability.
The DWBA model can be written as:
|
| (2) |
/
where is the acoustic wavelength, subscript 1 refers to the surrounding medium and subscript 2 to the object scattering the sound, 
is the position vector along the body axis, ac is the cross-sectional radius of the cylinder, 
and 
are related to the density and sound-speed contrasts, J1 is a Bessel function of the first kind of order 1, and ßtilt is the angle between the cylinder axis and the incident wave. The TS is calculated by
|
| (3) |
|
Preliminary TS measurements using a physical model and shrimps |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
Here we review the results from TS measurements using a physical model and freshwater shrimps to verify our TS-measurement method (Amakasu et al., 2003). The physical model was a copper, finite-length, straight cylinder whose TS-pattern was calculated theoretically by the straight-cylinder method (Stanton, 1988). The copper cylinder was suspended by a monofilament line and the orientation was varied. As shown in Figure 3, measured TS agreed with the results of the straight-cylinder model, confirming its accuracy.
|
Next, the TS measurements of shrimps were conducted to verify whether the TS-pattern of small animals such as krill could be measured by this method. We measured the TS-patterns of six live river shrimp (Palaemon paucidens) whose shape and material property were similar to those of krill. It was difficult to keep the shrimp in the beam due to their benthic nature. Therefore, their pleopods were removed and the shrimps were suspended by a monofilament line to maintain their position in the beam. Also their orientations were manipulated by the line. Figure 4 shows the experimentally measured TS and the TS-patterns by the DWBA model of the shrimps. The measured TS-patterns of the shrimps were in reasonable agreement with the theoretical TS-patterns. We had confirmed, therefore, that our measurement systems were able to measure precisely the TS-patterns of small animals.
|
|
TS-pattern of Antarctic krill |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
The TS measurements were conducted aboard the research and training vessel "Umitaka-maru" of the Tokyo University of Marine Science and Technology in February 2003 during a Southern Ocean survey. Antarctic krill samples were collected by larva-net tows by night, and the TS-pattern measurements of 12 live Antarctic krill (Table 2) were made when the ship was relatively still.
|
Figure 5 shows the experimentally measured TS (circles) and TS-patterns by the DWBA model (solid line). The shape used in the model was obtained for each animal from the video image when the animal showed its side aspect to the video camera. All shapes used in the model are shown in Figure 6 and are typical of those observed during each experiment. The density and sound-speed contrasts were 1.0357 and 1.0279, respectively (Foote, 1990; Foote et al., 1990), for all animals. Hereafter, these parameters were consistently used for the model in this study. The sound-speed in seawater was calculated for each experiment, as shown in Table 2.
|
|
We see from Figure 5 that the measured TS of most of the animals was generally in good agreement with the DWBA model on the main lobe, but that it was not so in the side-lobe regions. The TS measurements of Animals 2 and 4 were higher than the model predictions at tilt angles between 15° and 75° away from the main lobe. For Animals 1, 5, 7, 8, 11, and 12, although there was little information at large tilt angles, the measurements were similar to those of Animals 2 and 4. Many TS measurements were made at angles between –45° and 45° for Animals 5 and 7. For these measurements, there was general agreement between the measurements and the model predictions, but the variations of the measurements exceeded 10 dB, and were particularly large near 0°. Although the measurements for Animal 3 were slightly smaller than the model predictions, the TS-patterns were similar between the measurements and the model. As Animal 6 was entangled in the monofilament line and showed its ventral aspect to the transducer, most measurements were made from ventral aspect. Also, the TS-pattern was broad because the abdomen was severely bent. For Animal 9, the measured TS agreed with the model predictions. This animal was female with a swollen cephalothorax. Animal 10 was the shortest, with a length of 30.3 mm. The few measurements between –15° and 15° were all larger than the model predictions.
In summary, although the measurements were generally in good agreement near the main lobe of the DWBA model, they were larger than the model predictions at other angles or away from the main lobe. The experimental data of McGehee et al. (1998) at 120 kHz also showed a similar discrepancy.
|
Cause of TS discrepancy between measurement and model |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
The cause of the TS discrepancy between the measurements and the DWBA model predictions could stem from the measurement system, the method of approach, or physical changes in the body of the animal.
We investigated the echo signal and noise levels processed by the coherent subtraction in the Animal 4 data, which showed great TS discrepancy. The signal-to-noise ratio (SNR) was 10 dB when the TS was –85.5 dB, so SNR would be 0 dB if TS was –95.5 dB. We see from this result that the SNR was not small because most TS measurements TS were larger than –80 dB. There might be, however, some possibility that the coherent subtraction did not completely cancel out the reverberations in the side-lobe regions where the echo level was small.
Since the position angle of each animal measured by the split-beam method was 2° on average and these animals were not far from the beam axis, the scattering must be mostly from the dorsal aspect. Therefore, no large scattering components from the lateral aspect confounded these measurements.
Next, we examined the bending of the abdomen, scattering from pleopods, complexity of shape, non-homogeneity of material properties, and the effects of the carapace as causes of the TS discrepancy, some of which were suggested by Demer and Conti (2003).
Bending of abdomen
Using the DWBA model, we examine the variations in TS-pattern attributable to the bending of the abdomen. First, we made various bent shapes by modifying the shape of 58 mm Antarctic krill shown in Figure 2a of Kils (1981) by referencing to a photograph [cover page of Nature, 345(6273) 1990] of some Antarctic krill with bent abdomens. These shapes were expressed by a function that expresses the dorsal line of the abdomen seen from the side (Figure 7). The dorsal line of the cephalothorax is considered as a straight line because it does not bend. Finally, the dorsal line of the abdomen was expressed by the following formula by analysing the various shapes:
|
| (4) |
is the degree of bending and x is greater than the cephalothorax length of 22 mm. The dorsal line of the abdomen is calculated until the sum of the length of cephalothorax and abdomen is the total length 58 mm.
|
Ten patterns of the bent shape were made using Equation (5) (Figure 8), changing the
from 0 to 2.6 such that the distance between the ends of the neighbouring dorsal lines was equal, given the bending number B from 1 to 10, as shown. The radius and coordinates of each cylinder element are required for the DWBA model. The radius was read from the shape shown in Figure 2a of Kils (1981). The centre coordinates of each cylinder element were obtained by subtracting the radius from the dorsal line. In estimating TS-patterns for the shape which length is different, the radius and coordinates were adjusted proportionately by the ratio of the length.
|
Figure 9 shows the TS-patterns of the DWBA model predictions at 70 kHz for 30-, 40-, and 50-mm long animals as examples. The sound-speed in seawater used in the model calculation was 1451 m s–1, the average value at the time of the measurements. With increased abdominal bending, the main lobe of the TS-pattern becomes broad, the maximum TS shifts to negative tilt, and the side-lobe levels increase. Above 30°, however, the effects of abdominal bending were slight.
|
Scattering from the pleopods
It is expected that scattering from the pleopods may influence the scattering from the body, because Antarctic krill have five pairs of pleopods, and are always moving them. In order to investigate the effects of their scattering, the TS-pattern including scattering from the pleopods was computed by the DWBA model. The shape of Animal 4 (Figure 10) was used because its measurements agreed at the main lobe but there was a large TS discrepancy in the side-lobe regions (Figure 5). The radius and coordinates of cylinder slices for the pleopods were read from the video image when the animal showed its side aspect to the video camera. Since the pleopods were aligned in pairs, they were placed symmetrically to the body centre line. In the model calculation, the frequency was 70 kHz and the sound-speed in seawater was 1451.4 m s–1 (Table 2).
|
The scattering amplitude is written in terms of the separate contributions of the scattering from the two parts:
|
| (5) |
Figure 11 shows the TS-pattern which includes the scattering from the pleopods by the DWBA model, the measured TS, and the TS-patterns that do not include the scattering from the pleopods by the DWBA model. By adding the contribution of the pleopods, the peak level of the side lobe at about 30° increased by about 4 dB, and the peak level of the main lobe decreased by about 2 dB. Although the measurements are still larger than the model predictions, this change of TS-pattern is interesting, and it seems likely that the TS discrepancy is at least partly explained by the existence of the pleopods. To examine the effects of the scattering from the pleopods in detail we should consider their movement in a model computation, but this is not easy.
|
Other plausible causes
The fact that Antarctic krill have a more complex shape than combined cylinder slices with varying radii may be one of the causes of the TS discrepancy (Demer and Conti, 2003). Lavery et al. (2002) predicted the TS-pattern of decapod shrimps by the DWBA-based volume-integral model using high-resolution, three-dimensional digitizations of animal shape obtained through the use of computerized tomography. The experimental data again were not in good agreement with the predictions of the DWBA-based volume-integral model in the side-lobe regions. This suggests that complexity of shape is not a decisive cause of the TS discrepancy.
It is clear from the previous studies (Stanton and Chu, 2000; Lavery et al., 2002) that the TS-pattern is changed by the variations of the density and sound-speed contrast along the body axis. Stanton and Chu (2000) conducted simulations for inhomogeneity of the contrasts along the body axis, and reported that the contrast variations tended to give several decibel rises in the side-lobe region. This can be considered as one of the causes of the TS variability, but the density and sound-speed contrasts along the body axis are not really known yet. Although the carapace of krill is hard and may influence scattering, its effects should not be large because its thickness is much less than the wavelength.
Considering all of the above, it has to be said that a dominant factor was not found. It may be reasonable to consider that the TS discrepancy results from a combination of several factors, some of which were considered above. Furthermore, although the phase variability stemming from the influence of noise (Demer and Conti, 2003) may be a dominant factor in the TS discrepancy, further empirical or theoretical investigation will be necessary to clarify this and to quantify its contribution.
|
Conclusions |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
A new TS-measurement method using a split-beam echosounder in a small tank was applied to Antarctic krill, and we obtained TS-patterns of 12 live Antarctic krill at 70 kHz. Hitherto there have been no measurements at this frequency.
We confirmed that the measured TS agreed well with the DWBA model in the main-lobe region. We also found that the DWBA model cannot fully explain, but can partially explain, the TS discrepancy in the side-lobe regions. Although it is not clear whether the cause of this discrepancy was the same, the phenomenon is similar to the results previously reported by McGehee et al. (1998).
Examination of the causes of TS discrepancy from several aspects suggested that it is probably caused by a combination of factors such as the effect of noise not subtracted by the coherent subtraction, scattering from the pleopods, and abdominal bending. No factor was decisive, however. The contribution of each factor may become clear by future TS measurement of Antarctic krill under complete orientation control, in which the pleopods are cut or specimens are anaesthetized.
|
Acknowledgements |
|---|
We thank Takashi Ishimaru and Masato Moteki of Tokyo University of Marine Science and Technology and the captain, officers, crew, and cadets of the research and training vessel "Umitaka-maru" of Tokyo University of Marine Science and Technology for their cooperation. Thanks are also due to David Demer and Stéphane Conti, Southwest Fisheries Science Centre, USA for their careful review of the manuscript and useful comments. The experimental work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. Participation of Kazuo Amakasu to present this study in UT '04 held in Taiwan was supported by the Inoue Foundation for Science.
|
References |
|---|
|
TopIntroductionMethodsPreliminary TS measurements...TS-pattern of Antarctic krillCause of TS discrepancy...ConclusionsReferences |
|---|
-
Amakasu K., Furusawa M., Fan C. (2003) A method for target-strength measurement of small-size animals by the split-beam method in a small tank. Nippon Suisan Gakkaishi 69:575–583 (in Japanese).
Chu D., Foote K.G., Stanton T.K. (1993) Further analysis of target-strength measurements of Antarctic krill at 38 and 120 kHz: comparison with the deformed-cylinder model and inference of orientation distribution. Journal of the Acoustical Society of America 93:2985–2988.[CrossRef][Web of Science]
Demer D.A. and Conti S.G. (2003) Reconciling theoretical versus empirical target strengths of krill: effects of phase variability on the distorted-wave, Born approximation. ICES Journal of Marine Science 60:429–434.
Ding L. (1997) Direct laboratory measurement of forward scattering by individual fish. Journal of the Acoustical Society of America 101:3398–3404.[CrossRef][Web of Science]
Foote K.G. (1990) Speed of sound in Euphausia superba. Journal of the Acoustical Society of America 87:41405–1408.[CrossRef][Web of Science]
Foote K.G., Everson I., Watkins J.L., Bone D.G. (1990) Target strengths of Antarctic krill (Euphausia superba) at 38 and 120 kHz. Journal of the Acoustical Society of America 87:16–24.[CrossRef][Web of Science]
Foote K.G., Knudsen H.P., Vestnes G., MacLennan D.N., Simmonds E.J. (1987) Calibration of acoustic instruments for fish-density estimation: a practical guide. ICES Cooperative Research Report 144: 69 pp.
Furusawa M., Miyanohana Y., Ariji M., Sawada Y. (1994) Prediction of krill target strength by liquid, prolate-spheroid model. Fisheries Science 60:261–265.[Web of Science]
Furusawa M., Miyanohana Y., Sawada K., Takao Y. (1995) Calibration manual for quantitative echo sounders. Technical Report of the National Research Institute of Fisheries: Engineering, Fishing Boats and Instruments 15: pp. 9–37 (in Japanese).
Hewitt R.P. and Demer D.A. (1991) Krill abundance. Nature 353:310.
Hewitt R.P., Watkins J.L., Naganobu M., Tshernyshkov P., Brierley A.S., Demer D.A., Kasatkina S., Takao Y., Goss C., Malyshko A., Brandon M.A., Kawaguchi So., Siegel V., Trathan P.N., Emery J.H., Everson I., Miller D.G.M. (2002) Setting a precautionary catch limit for Antarctic krill. Oceanography 15:326–33.
Kils U. (1981) Swimming behaviour, swimming performance and energy balance of Antarctic krill, Euphausia superba. BIOMASS Scientific Series 3: 122 pp.
Lavery A.C., Stanton T.K., McGehee D.E., Chu D. (2002) Three-dimensional modelling of acoustic backscattering from fluid-like zooplankton. Journal of the Acoustical Society of America 111:1197–1210.[CrossRef][Web of Science][Medline]
McGehee D.E., O'Driscoll R.L., Martin Traykovski L.V. (1998) Effects of orientation on acoustic scattering from Antarctic krill at 120 kHz. Deep-Sea Research II 45:1273–1294.[CrossRef]
Pauly T. and Penrose J.D. (1998) Laboratory target-strength measurements of free-swimming Antarctic krill (Euphausia superba). Journal of the Acoustical Society of America 103:3268–3280.[CrossRef][Web of Science]
Sawada K., Ye Z., Kieser R., McFarlane G.A., Miyanohana Y., Furusawa M. (1999) Target-strength measurements and modelling of walleye pollock and Pacific hake. Fisheries Science 65:193–205.[Web of Science]
Stanton T.K. (1988) Sound scattering by cylinders of finite length. II. Elastic cylinders. Journal of the Acoustical Society of America 83:64–67.[CrossRef][Web of Science]
Stanton T.K. and Chu D. (2000) Review and recommendations for the modelling of acoustic scattering by fluid-like, elongated zooplankton: euphausiids and copepods. ICES Journal of Marine Science 57:793–807.
Stanton T.K., Chu D., Wiebe P.H. (1998) Sound scattering by several zooplankton groups. II. Scattering models. Journal of the Acoustical Society of America 103:236–253.[CrossRef][Web of Science][Medline]
Stanton T.K., Chu D., Wiebe P.H., Clay C.S. (1993) Average echoes from randomly oriented, random-length finite cylinders: zooplankton models. Journal of the Acoustical Society of America 94:3463–3472.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Furusawa and K. Amakasu The analysis of echotrace obtained by a split-beam echosounder to observe the tilt-angle dependence of fish target strength in situ ICES J. Mar. Sci., November 13, 2009; (2009) fsp246v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||