Skip Navigation

ICES Journal of Marine Science: Journal du Conseil 2006 63(5):952-955; doi:10.1016/j.icesjms.2005.12.011
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Dunford, A. J.
Right arrow Articles by Macaulay, G. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dunford, A. J.
Right arrow Articles by Macaulay, G. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2006 International Council for the Exploration of the Sea

Progress in determining southern blue whiting (Micromesistius australis) target strength: results of swimbladder modelling

Adam J. Dunford* and Gavin J. Macaulay

National Institute of Water and Atmospheric Research Private Bag 14-901, Kilbirnie, Wellington, New Zealand

*Correspondence to A. J. Dunford: tel: +64 4 386 0300; fax: +64 4 386 0574. e-mail: a.dunford{at}niwa.cri.nz.

Southern blue whiting target strength (TS) results from Kirchhoff modelling of swimbladder casts scanned using a hand-held 3D laser scanner are presented. The data are compared with the relationship between TS and fish length used for New Zealand stock-assessment surveys; TS = 21.8 log10(fork length) – 72.8, at 38 kHz. This relationship has its origins in the relationship used for blue whiting (Micromesistius poutassou) in the northern hemisphere, and is based on measurements on juvenile cod (Gadus morhua). The results indicate that the blue whiting relationship is not appropriate for southern blue whiting, and suggest a much steeper slope, with TS = 38 log10(fork length) – 97, at 38 kHz. Sensitivity analyses indicate that further investigations of swimbladder tilt-angle distribution and swimbladder volume are unlikely to provide evidence to support the use of the blue whiting relationship for southern blue whiting.

Keywords: acoustic target strength, Kirchhoff modelling, southern blue whiting, swimbladder

Received 25 May 2005; accepted 3 December 2005.


   Introduction
Top
 Introduction
Methods
 Results and discussion
 Appendix
 References
 
Southern blue whiting (Micromesistius australis) support significant commercial fisheries around South America and New Zealand. During August and September, they form monospecific spawning aggregations in depths of 250–500 m on the Campbell and Bounty Plateau in the Subantarctic waters off southeastern New Zealand. Estimates of relative abundance from acoustic surveys of these aggregations are one of the primary stock-assessment tools for southern blue whiting (SBW) in New Zealand. Although the absolute value of target strength (TS) is not critical for a time-series of relative abundance estimates, changes in the slope of the relationship between TS and fish length will affect the time-series if the distribution of fish lengths changes over time.

The relationship between TS and length for SBW used in New Zealand is the same as that used for blue whiting (Micromesistius poutassou; BW) in the northern hemisphere. The BW TS–length relationship is also used on Argentine (Madirolas, 1999) and Chilean (S. Lillo, IFOP, Chile, pers. comm.) acoustic surveys of SBW. This relationship is based on measurements on juvenile cod (Gadus morhua; Nakken and Olsen, 1977) which were later re-analysed by Foote (1980) and taken to be representative of BW. Northern hemisphere in situ split-beam measurements of BW, which have begun recently, indicate that the BW TS is higher than previously thought (Godø et al., 2002), and further in situ work is planned to allow a more definitive relationship to be obtained (O. Godø, Institute of Marine Research, Bergen, pers. comm.). Here we review swimbladder-modelling data collected for SBW in New Zealand waters and present a new TS–length relationship for SBW.

Swimbladder modelling is a useful adjunct to in situ measurements because it allows investigation of how variables such as swimbladder tilt angle impact on TS, and also gives results that are independent of in situ data. In this paper the Kirchhoff modelling technique (Foote, 1985; Medwin and Clay, 1998) is used to compute tilt-averaged TS values from scanned swimbladder casts. The contribution of the fish body to the target strength is neither estimated nor allowed for in this manuscript, for three reasons. First, for fish with a gas-filled swimbladder, the contribution is relatively small (estimated at 5–10% of the backscattered energy; Foote, 1985). Second, the swimbladder injection technique and subsequent scanning were devised as a fast and convenient method to obtain TS estimates, but obtaining the shape and size of the fish body made the process considerably longer and more involved. Finally, injecting the swimbladders is a destructive technique, and fish-body data would need to be obtained before injecting the fish. Unfortunately, however, the 3D scanner was unavailable during the voyages that collected the swimbladder casts.

Early in situ results and some earlier Kirchhoff modelling of sectioned SBW plaster swimbladder casts have previously been published by McClatchie et al. (1998). The swimbladder-modelling data were calculated by slicing the swimbladder casts, digitizing the two-dimensional slices, and reconstructing the 3D shape for the calculation of target strength. The slicing and reconstruction process were manually intensive, and the software for calculating the TS from the Kirchhoff approximation model had several coding problems which have been corrected since those results were produced. Because of this, the swimbladder-modelling results of McClatchie et al. (1998) are not included in this analysis.


   Methods
Top
 Introduction
 Methods
Results and discussion
 Appendix
 References
 
Swimbladder casts for TS modelling were collected during the 1997, 2001, and 2002 SBW acoustic surveys off southern New Zealand. To assess the amount of resin to inject into each swimbladder, a number of fish were weighed in water, taking care to remove any trapped air bubbles. This weight, corresponding to the force required to make the fish neutrally buoyant, gives the volume of the swimbladder, and hence the volume of resin required. In all, 74 fish with lengths in the range 19–58 cm were measured in this way, and a relationship between fish length and buoyancy was derived and used to calculate the amount of resin to inject into each fish.

Swimbladder casts were collected from a second group of 78 fish (length range 16–58 cm) by injecting the swimbladder with epoxy resin while still in the body cavity of the fish and dissecting it out after the epoxy had cured. The casts were scanned using a hand-held, laser, 3D scanner (Polhemus, 2000), which produced a triangular mesh representing the swimbladder surface. Five casts were rejected because they contained scanning artefacts (large holes) that affect the computed TS. Using the Kirchhoff-approximation model (Foote, 1985; Medwin and Clay, 1998), the TS at 38 kHz was calculated at angles of –40° to +40° in steps of 1°. These data were convolved with a fish tilt-angle distribution having a mean of 0° and standard deviation of 15° (McClatchie et al., 1998) to obtain a tilt-averaged TS estimate for each fish.

A new TS–length relationship, of the form <TS> = mlog10(l) + c, where l is the fork length in cm and <TS>, m, and c are tilt-averaged target strength, slope, and intercept, respectively, in dB re 1 m2, was calculated using a least-squares regression. As recommended by McClatchie et al. (2003), no attempt was made to force the relationship slope through 20.

The two main unknowns in modelling swimbladders are the tilt-angle distribution and the inflation level. The sensitivity of the tilt-averaged TS to changes in the tilt distribution was investigated by re-computing the TS for each swimbladder using a range of tilt-angle distributions. The mean and the standard deviation of the tilt-angle distribution were varied from –10° to 10° and 5° to 25°, respectively, in steps of 5°. A length-to-TS regression was then fitted to the re-computed TS for each tilt distribution using the Matlab least-squares fitting function (The Mathworks Inc., 2000).

A similar process was used to investigate the effect of inflation levels on TS. All three dimensions (length, width, and height) of each swimbladder were scaled by a constant factor. The range of size-scaling factors was determined by comparing the volume required for buoyancy and the actual swimbladder volume. This suggested that the swimbladder volume and the "true" volume could be different by up to a factor of two, so a size-scaling factor range of Formula to Formula was used. For each size-scaling factor, the TS of all swimbladders was re-computed, using the tilt-angle distribution of McClatchie et al., (1998; mean 0°, s.d. 15°), and a length-to-TS regression obtained.


   Results and discussion
Top
 Introduction
 Methods
 Results and discussion
Appendix
 References
 
The swimbladder-modelling results for SBW are not consistent with the BW relationship (Figure 1). These results are tabulated in the Appendix to provide a reference for comparison with other data sets. The regression fit to our swimbladder data was <TS> = 38log10(l) – 97, compared with the BW relationship of <TS> = 21.8log10(l) – 72.8.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 Southern blue whiting swimbladder-modelling data. The data points are tilt-averaged target strength (<TS>) calculated using a tilt-angle distribution of mean 0° and s.d. 15°. The solid line shows the blue whiting relationship currently used for southern blue whiting stock assessment and the dotted line shows the relationship obtained from the swimbladder data.

 
Figure 2 shows the in situ data of McClatchie et al. (1998) and Argentine in situ data for SBW (A. Madirolas, pers. comm.) which agree reasonably well with our swimbladder-modelling relationship. In addition, Figure 2 shows recent BW in situ data (Forbes, 1985; Godø et al., 2002; Heino et al., 2003), which do not agree with our relationship.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Southern blue whiting TS relationships compared with southern blue whiting and blue whiting in situ data. The solid line shows the blue whiting relationship currently used for southern blue whiting stock assessment and the dotted line shows the relationship obtained from the swimbladder data.

 
A number of factors can affect TS and may not be adequately captured by swimbladder modelling, for example, changes in TS with gonad stage or feeding, changes with depth, and changes caused by behaviour. For a species with an air-filled swimbladder, such as SBW, many of these effects can be modelled by changing either the tilt distribution used in calculating the TS or the volume of the swimbladder. Changing the tilt-angle distribution has little effect on the slope of the regression (Figure 3), so regardless of what distribution is finally adopted, it is likely that the "true" relationship will have a higher slope than the BW relationship. Similarly, scaling of swimbladder dimensions has little effect on the slope of the relationship (Figure 4).


Figure 3
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3 Southern blue whiting modelling data showing the effect of changing the tilt-angle distribution used for computation. The thin solid line shows the blue whiting relationship currently used for southern blue whiting stock assessment and the dotted lines show the regression to the re-computed data. The thicker solid line is for a mean/s.d. of 0°/15°.

 


Figure 4
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4 Southern blue whiting modelling data showing the effect of varying inflation levels. The swimbladders were "inflated" by scaling in each dimension by a constant factor. The thin solid line shows the blue whiting relationship currently used for southern blue whiting stock assessment and the dotted lines show the regression to the re-computed data for each scaling factor. The thicker solid line is for a scaling factor of 1.

 
The major sources of error in swimbladder modelling arise from the difference between the "true" swimbladder tilt-angle distribution and inflation level and those used in the calculations. As the sensitivity analyses have shown, it is unlikely that further investigations will provide evidence to support the use of the BW relationship for SBW. For an average SBW of 45 cm (Paul, 2000) the TS from the BW relationship and that based on the swimbladder data are –36.8 dB and –34.2 dB, respectively. For a population with mean length 45 cm, adoption of the swimbladder-modelling relationship would reduce acoustic biomass estimates by approximately 45%.


   Appendix
Top
 Introduction
 Methods
 Results and discussion
 Appendix
References
 


View this table:
[in this window]
[in a new window]

 
 

   Acknowledgements
 
We thank the scientific staff and the officers and crew of RV "Tangaroa", who supported the seagoing aspects of this project for many years. Stuart Hanchet, Sam McClatchie, and Richard O'Driscoll are thanked for their comments on earlier drafts of the manuscript, and Paul Grimes and Alan Hart for their assistance with the swimbladder scanning. Provision of data by Adrian Madirolas of Instituto Nacional de Investigación y Desarrollo Pesquero in Argentina is gratefully acknowledged. The research was funded in part by the New Zealand Ministry of Fisheries through projects MDBW03, MDBW02, MBW702, SBW9701, SBW9801, SBW9901, SBW2000/02, and SBW2001/02.


   References
Top
 Introduction
 Methods
 Results and discussion
 Appendix
 References
 

    Foote K.G. (1980) Averaging of fish target-strength functions. Journal of the Acoustical Society of America 67:504–515.[CrossRef][Web of Science]

    Foote K.G. (1985) Rather-high-frequency sound scattering of swimbladdered fish. Journal of the Acoustical Society of America 78:688–700.[CrossRef][Web of Science]

    Forbes S. T. (1985) Progress in dual-beam target-strength measurement on herring and blue whiting. ICES CM/1985/B: 22. 5 pp.

    Godø O. R., Heino M., Søiland H., Alvarez J., Dahl M., de Lange J., Gullaksen O., Ø. Tangen Ø., Torkelson T. (2002) Blue-whiting survey during spring 2002. Working document of the ICES Northern Pelagic and Blue Whiting Fisheries Working Group. 27 pp.

    Heino M., Godø O. R., Søiland H., Nygaard J. E., Alvarez J., Drivenes L., de Lange J., Gullaksen O., Tangen Ø., Torkelson T. (2003) Blue-whiting survey during spring 2003. Working document of the ICES Northern Pelagic and Blue Whiting Fisheries Working Group. 40 pp.

    Madirolas A. (1999) Acoustic surveys on the southern blue whiting. INIDEP Documento Cientifico 5:81–93.

    McClatchie S., Macaulay G.J., Coombs R.F. (2003) A requiem for the use of 20log10 length for acoustic target strength with special reference to deep-sea fishes. ICES Journal of Marine Science 60:419–428.[Abstract/Free Full Text]

    McClatchie S., Macaulay G.J., Hanchet S.M., Coombs R.F. (1998) Target strength of southern blue whiting (Micromesistius australis) using swimbladder modelling, split beam and deconvolution. ICES Journal of Marine Science 55:482–493.[Abstract/Free Full Text]

    Medwin H. and Clay C.S. (1998) Fundamentals of Acoustical Oceanography(Academic Press, Boston).

    Nakken O. and Olsen K. (1977) Target-strength measurements of fish. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer 170:52–69.

    Paul L.J. (2000) New Zealand FishesReed Publishing (NZ) Ltd.

    Polhemus. (2000) User manual for the FastSCAN hand-held laser scanner. Polhemus Inc.

    The Mathworks Inc. (2000) MATLAB: The Language of Technical Computing: Computation, Visualization, Programming: Using MATLAB. Version 6. The Mathworks Inc.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Dunford, A. J.
Right arrow Articles by Macaulay, G. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dunford, A. J.
Right arrow Articles by Macaulay, G. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?