© 2005 International Council for the Exploration of the Sea
In situ target-strength measurement of young hairtail (Trichiurus haumela) in the Yellow Sea
a College of Marine Life Science, Ocean University of China, Qingdao, China
b Key Laboratory for Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, 106 Nanjing Road, Qingdao 266071, China
*Correspondence to X. Zhao: Key Laboratory for Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, 106 Nanjing Road, Qingdao 266071, China; tel: +86 532 858 353 63; fax: +86 532 858 115 14. e-mail: zhaoxy{at}ysfri.ac.cn.
The target strength of hairtail (Trichiurus haumela) in the Yellow Sea was measured in situ with a 38 kHz, split-beam echosounder on 2 January 2001. The fish measured were of the 2000 year class, its anal length ranged from 62 to 115 mm, with a mean of 89.8 mm. The mean target strength of these young hairtail was estimated to be –49.2 dB, with a 95% confidence interval of (–49.4, –49.0) dB. This provided a rare and useful reference for the acoustic-abundance estimation of hairtail.
Keywords: hairtail, target strength, Yellow Sea
Received 29 May 2003; accepted 30 July 2005.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Hairtail [Trichiurus haumela (Forsskål)] is one of the commercially and ecologically most important fish species in the East China Sea and the Yellow Sea (Luo, 1991). It is a swimbladder-bearing, semi-demersal fish that performs fairly distinct diurnal vertical migrations. It concentrates in a narrow layer close to, but often a few metres above, the bottom during daytime (own observation), and disperses in the pelagic during night-time. This makes the fish a suitable target for echo-integration surveys. However, hairtail are often mixed with many other fish of non-negligible quantity, which renders the in situ measurement of the target strength of this fish a very difficult task.
Hairtail has an extremely elongated and deeply compressed body shape; its unusually thin and long tail is easily broken, which leads to the measurement of its anal length instead. Lacking direct target-strength (TS) measurements, Ona (1987) suggested the following TS to anal-length (la) relation for hairtail:
|
| (1) |
During an acoustic/trawl survey on the recruitment status of important fish populations in the Yellow Sea in winter 2000/2001, hairtail echo registrations suitable for in situ target-strength measurement were encountered. The results are discussed in this paper.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Measurement site
The survey was conducted aboard RV "Bei Dou", a 56.2 m stern trawler designed for acoustic and trawl surveys. Target-strength measurement of hairtail was carried out in the western part of the Yellow Sea (Figure 1), close to a traditional spawning ground in Haizhou Bay (Luo, 1991).
|
Biological sample collection
During the survey, at about 18:30 on 2 January 2001, dispersed single-fish echo traces (Figure 2) were found in the water column from 25 m below the sea surface to a few metres above the bottom at about 52-m depth. In order to identify the biological origin of the echo traces, a sampling haul was carried out. In this particular instance, a bottom trawl, instead of midwater trawl, was used to test the effectiveness of the bottom trawl as a pelagic sampling device at shallow depth.
|
The trawl was monitored via a SCANMAR trawlsounder (model TS150). At a speed of 3.2 knots, the trawl was initially towed pelagically at 25-m depth by paying out only 100 m of warp; every 20 m additional warp was paid out after the net had been stabilized for about 10 min at each sampling depth. The trawl finally stabilized on the bottom for 10 min at a warp length of 190 m. The tow lasted for 56 min. The start and end positions of the tow were 34°32.4'N 122°15.0'E and 34°32.2'N 122°11.4'E, respectively.
The trawl was made up of four panels, with a stretched mesh size of 200 mm in the forward section and 14 mm in the codend. The vertical opening of the trawl ranged from 7.7 to 8.2 m when towed pelagically, and from 5.9 to 6.1 m when settled on the bottom; its wingspread was about 25 m according to previous and later measurements.
The trawl sample was sorted according to the routine procedure. For hairtail, a subsample of 110 fish was measured to the nearest millimetre (anal length) to give the length distribution.
Echo-trace data acquisition
When the haul was on deck, the survey was adjourned for target-strength measurement. The ship sailed back to the start position of the haul and then turned back to the end position again at a speed of about 10 knots. The echo-trace data were collected during these two runs with a SIMRAD EK500, split-beam echosounder. The echosounder was calibrated prior to the survey on 26 December 2000 using the LOBE program (V5.xx, 30 October 1995) provided by SIMRAD. The calibration results and other important instrument settings are given in Table 1.
|
Data analysis
The raw echo-trace data were logged via the serial port of the echosounder and stored on a PC by PROCOMM (V2.4, Datastorm Technologies Inc., 1986). They were processed using a dedicated Pascal program (Zhao, 1996). First, the echogram was carefully checked and only those runs with a constant vessel speed of about 10 knots and clear single-fish echo traces were selected for further treatment. The corresponding sub-data set was obtained by specifying the selected time intervals and range limits through a subroutine of the program. Data were then reformatted into the standard case-by-variable format and imported into SYSTAT Inc (1992) for final analysis. In all cases, the averaging process was done in the linear domain (scattering cross-section) and then converted into the logarithmic domain (TS).
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Biological samples
Hairtail dominated the catch both in number and in weight (Table 2). Figure 3 shows the length distribution of the hairtail sample. The anal length ranged from 62 to 115 mm, with a mean of 89.8 mm and standard deviation of 9.7 mm. They were of the 2000 year class, according to the growth characteristics of this fish (Luo, 1991).
|
|
Anchovy (Engraulis japonicus) was the second most common species in the catch, its total length ranging from 75 to 98 mm. Other fish in the haul were of negligible amount (Table 2). Some typical bottom-dwelling fish and small shrimp and crab were also present. They were most likely caught at the end of the tow when the trawl was on the bottom, and are not included in Table 2.
Target strength of hairtail
Figures 4 and 5 show, respectively, the target-strength histogram of the echo data selected from 25- to 45-m depth layer and the layer deeper than 45 m (up to 53 m). It can be seen from Figure 4 that the target-strength distribution has a long left tail, with a distinct mode at –48 dB; a smaller second mode at –59 dB is also visible. The target-strength distribution of the echoes close to the bottom (Figure 5) is relatively messy, with more large echoes than those observed in the upper layer of the water column (Figure 4).
|
|
Since most of the best single-fish echo traces were concentrated in the 25–45 m layer (Figure 2), and also to avoid further contamination attributable to other organisms close to the bottom, only data in the depth range 25–45 m were used for the hairtail target-strength estimate. The mean target strength of hairtail was thus estimated to be –49.2 dB, with a 95% confidence interval of (–49.4, –49.0) dB. The mean depth of the echo traces was 38.2 m.
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Target strength of hairtail
The target strength of a fish is a key parameter for quantitative acoustic fish-abundance estimation. For some fish, however, this parameter is often not available or even not easily obtainable. Hairtail is such a fish. It is either mixed with other fish in non-negligible quantity or aggregated in too high density and not resolvable as a single target. In fact, I have not found any references in the literature on the target-strength measurement of this particular fish.
The mean target strength of 9.0 cm (anal length) hairtail reported in this paper was –49.2 dB, with a 95% confidence interval of (–49.4, –49.0) dB. When expressed in the common TS = 20 log 1a + b20 formula, it becomes:
|
| (2) |
Sources of error
For the target-strength measurement of hairtail reported in this paper the estimate was still open to several sources of error, even though the data were carefully selected. First, the minimum target strength (–60 dB) used for single-fish, echo-trace data collection was a little too high, as can be seen from the truncated left tail of the target-strength distribution (Figure 4). This might lead to a slightly higher TS estimate.
Second is the long recognized multiple-target problem (Ona and Røttingen, 1986; Soule et al., 1995) although the performance of the single-target detection algorithm of the new echosounder software has been greatly improved (Soule et al., 1997). The maximum sA for the 30–40 m layer was 21 m2 nautical mile–2, corresponding to a fish density of less than 0.1 fish per sampled volume, or a multiple-target probability of less than 0.05 according to Ona (1999). The fish density in the 20–30 m layer was much less. Therefore, failure in the complete rejection of multiple targets might cause a positive but very small bias.
The third and probably the most complicated source of error is the contamination of hairtail echo data due to the presence of several other fish. Fortunately, anchovy was probably the only contaminant that was of practical significance in the depth range 25–45 m. According to Chen and Zhao (1990), the TS to fish-length relationship of anchovy is about TS = 20 log 1 – 73.2. The mean target strength of these 75–98 mm anchovy was therefore anticipated to be lower than that of hairtail, so the estimated mean target strength of hairtail might be negatively biased.
The extent of these biases, however, is very difficult to quantify. In the case of the truncated left tail, Foote et al. (1986) used a compensation procedure based on theoretical or simulated target-strength distribution; while in the case of contamination or mixed species, several statistical approaches have been used to extract those target-strength data that are believed to be connected with the fish in question (Foote et al., 1986; Barange et al., 1996; MacLennan and Menz, 1996). A common prerequisite to use of these approaches is that the target-strength distribution of the fish in question is known or assumed a priori. However, there is simply no a priori knowledge of the target-strength distribution of the hairtail on which to base an assumption. This is mainly due to the uncertainty about the behaviour of this fish in connection with the peculiarity of its body shape, behavioural pattern being an important influential factor in the determination of target strength (Foote, 1980). Moreover, because of the directive nature of sound scattering on the one hand, and the non-uniform, individual behavioural pattern of the fish on the other, the target-strength distribution of fish can be bimodal (Williamson and Traynor, 1984) or even multi-modal and quite variable (Barange and Hampton, 1994). Zhao (1996) showed that repeated measurements on a single free-swimming fish could result in multi-modal target-strength distribution, and warned that great care should be paid when trying to link different modes in the target-strength distribution directly to those of the fish size distribution.
Hairtail has a thin but long swimbladder (Figure 6). The ratio between the length of the main chamber of the swimbladder and the anal length of the fish is 0.48 according to Ona (1987). Therefore hairtail must be a directive scatterer at 38 kHz, so a target-strength distribution similar to that shown in Figure 4 can be considered as highly possible. In fact, multi-modal or even a skewed distribution pattern like that shown in Figure 4 has also been found with anchovy (X. Zhao, unpublished observation). Therefore, no attempt has been made to "correct" the target-strength estimate of hairtail at present if only because hairtail was the overwhelmingly dominant fish according to the trawl sample (Table 2). Possible bias associated with the selectivity of the sampling gear and its avoidance by fish is well known (Foote et al., 1986; Barange and Hampton, 1994; MacLennan and Menz, 1996), and will not be discussed further here.
|
Future work
The target-strength estimate given in this paper provides a valuable starting point for the acoustic-abundance estimation of hairtail not only in the Yellow Sea and East China Sea, but also in other parts of the world. Further efforts should be made to measure the target strength of this fish of different lengths and preferably at even purer catch composition, and, of course, using a lower minimum target strength for single-fish, echo-trace data collection.
|
Acknowledgements |
|---|
The captain and crew of the RV "Bei Dou" are thanked for their cooperation during the unusual sampling operation, and Yong Wang for his help during target-strength data collection. Dr E. Ona of the Institute of Marine Research, Bergen, Norway and two anonymous reviewers are thanked for their valuable comments. This work was carried out under the auspices of the National Fishery Resources Monitoring Program from the Ministry of Agriculture (HY126-02) and the National Key Basic Research Program from the Ministry of Science and Technology, P. R. China (G19990437).
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Barange M. and Hampton I. (1994) Influence of trawling on in situ estimates of Cape horse mackerel (Trachurus trachurus capensis) target strength. ICES Journal of Marine Science 53:225–232.[CrossRef]
Barange M., Hampton I., Soule M. (1996) Empirical determination of in situ target strength of three loosely aggregated pelagic fish species. ICES Journal of Marine Science 53:225–232.
Chen Y. and Zhao X. (1990) In situ target strength measurements on anchovy (Engraulis japonicus) and sardine (Sardinops melanostictus). Proceedings of International Workshop on Marine AcousticsBeijing, China(China Ocean Press, Beijing) pp. 329–332 388 pp.
Foote K.G. (1980) Effect of fish behaviour on echo energy: the need for measurements of orientation distributions. Journal du Conseil International pour l'Exploration de la Mer 39:193–201.
Foote K.G., Aglen A., Nakken O. (1986) Measurement of fish target strength with a split-beam echosounder. Journal of Acoustical Society of America 80:612–621.[CrossRef]
Luo B.Z. (1991) Hairtail. In Deng J.Y. and Zhao C.Y. (Eds.). Marine Fisheries Biology(Agriculture Press, Beijing) pp. 111–163 686 pp (in Chinese).
MacLennan D.N. and Menz A. (1996) Interpretation of in situ target-strength data. ICES Journal of Marine Science 53:233–236.
Ona E. (1987) Adjustment of the average target strength of hairtail (Trichiurus haumela). Internal note, Yellow Sea Fisheries Research Institute, Qingdao. 6 pp.
In Ona E. (Ed.). Methodology for target-strength measurements (with special reference to in situ techniques for fish and micronekton). (1999) ICES Cooperative Research Report, No. 235. 59 pp.
Ona E. and Røttingen E. (1986) Experiences using the ES400, split-beam echosounder, with special reference to the single-target recognition criteria. ICES CM 1986/B:38. 8 pp.
Soule M.A., Barange M., Hampton I. (1995) Evidence of bias in estimates of target strength obtained with a split-beam echosounder. ICES Journal of Marine Science 52:139–144.
Soule M.A., Barange M., Solli H., Hampton I. (1997) Performance of a new phase algorithm for discriminating between single and overlapping echoes in a split-beam echosounder. ICES Journal of Marine Science 54:934–938.
SYSTAT Inc. (1992) SYSTAT for Windows: Statistics Version 5 edn Evanston, IL. 636 pp.
Williamson N.J. and Traynor J.J. (1984) In situ target-strength estimation of Pacific whiting (Merluccius productus) using a dual-beam transducer. Journal du Conseil International pour l'Exploration de la Mer 41:285–292.
Zhao X. (1996) Target strength of herring (Clupea harengus L.) measured by the split-beam tracking method. M.Phil. thesis, Department of Fisheries and Marine Biology, University of Bergen. 103 pp.
Zhu D. and Iversen S.A. (1990) Anchovy and other fish resources in the Yellow Sea and East China Sea. Marine Fisheries Research (China) 11:1–143 (in Chinese with English abstract).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||