.

Fishery independent surveys:International North Sea herring acoustic survey

Fisheries acoustics

The penetration of sound in water is significantly greater than that of light. Active acoustic instruments which, by definition, transmit and receive sound waves are, therefore, capable of detecting fish or other objects far beyond the range of any visual system. The exploitation of this property is most evident in the sophisticated echolocation facilities of whales and dolphins (Au, 1993). Acoustic instruments, developed over the last century (Fernandes et al., 2001), are now essential requirements for any fishing vessel to determine the location of fish and the seabed. Similarly, acoustic applications are now widespread in fisheries science to assess the abundance, distribution, and behaviour of fish, plankton, and other marine organisms; they are also used in monitoring the performance of sampling gears (see Walsh et al., 2001).
The essential physical principles of the discipline are covered in a variety of specific fisheries acoustics texts (Forbes and Nakken, 1972; Cushing, 1973; Mitson, 1984), the latest of which (MacLennan and Simmonds, 1992) provides the most comprehensive treatment. The more general sciences of underwater acoustics and acoustical oceanography have been covered extensively by Urick (1983) and Medwin and Clay (1998) respectively.

The echosounder The basic tool in fisheries acoustics is the scientific echosounder (Fig. 1). This instrument produces an electrical signal which is converted by a transducer to an acoustic pulse or "ping". The transducer is mounted on a suitable platform, such as the hull of a ship, and so the "ping" is directed vertically downwards into the water column in a beam (typically of the order of 10°) which geometrically is the acoustical analogue of a beam of light from a torch. When objects such as fish are ensonified by the sound, part of the acoustic energy is reflected and received by the transducer as an echo which is then converted to electrical energy. The distance or range to the fish is obtained by timing the interval between transmission and reception (knowing the speed of sound in water, approximately 1500 m s-1). The energy is then amplified to compensate for the effects of geometrical spreading and absorption.
	Fig. 1: Principles of echosounding and echo integration. Echoes, reflected from the fish school and the seabed are displayed on the echogram. The intensity of the school echo is measured as the area backscattering coefficient (sA).

Abundance estimation

An echosounder will typically ping at a rate of one pulse per second. When calibrated, the absolute echo levels are quantified by averaging a number of transmissions using echo integration, yielding a quantity which is proportional to fish density according to the linearity principle (Foote, 1983). The calculated fish densities, obtained from survey vessels travelling along defined transects, are then interpolated and raised to the survey area to give an estimate of fish abundance (Simmonds et al., 1992). The acoustic properties of the fish must be known and these are obtained from species specific target strength (TS) relationships once the fish have been identified. Although echo characteristics may often be sufficient to identify fish to species (Reid, 2000), confirmation is obtained from trawl samples which also provide length, age, and maturity composition.
Within the ICES community, there are currently over 20 fish stocks for which acoustic assessments are carried out. Most of these are pelagic (midwater) species such as herring, sprat, sardine, and anchovy. The technique has traditionally not been suitable for the detection of demersal fish that occur very close to the bottom in the acoustic "dead-zone” (Mitson, 1983). This is, however, changing with improvements in seabed recognition and the application of correction factors (Ona and Mitson, 1996). Throughout the world, acoustic techniques are used in an even greater variety of fisheries, such as Antarctic krill (Hewitt and Demer, 1991) and the deep water acoustic surveys of orange roughy off Tasmania (Kloser, 1996).

Acoustic surveys in the North Sea

The idea for an acoustic survey for herring covering the entire North Sea dates back to the early 1950s when Parish (1953) first proposed an "..organised echo search..” to determine the distribution of the species (Figure 2). However, it was not until 1979, one year after the closure of the North Sea herring fishery, that the first acoustic survey was conducted (ICES, 1980). Six ships were employed covering the northwestern and west central North Sea and northern part of Division VIa. However, the results of this survey were so variable that the area coverage was reduced to the Orkney Shetland area the following year (1980). Coverage was restricted to the northern North Sea until 1988 when once again a number of surveys were carried out at sufficiently similar times to enable a combined estimate to be obtained (ICES, 1989). Since then the ICES international North Sea herring acoustic survey has taken place every year, with the participation of the United Kingdom (Scotland), the Netherlands, Norway, Germany, Denmark and, occasionally, Sweden and the Republic of Ireland. (see Bailey et al., 1998 for a review of the time series). Typically the survey covers the whole of the North Sea from late June to early August (Fig. 3).
	Fig. 2: (Reproduction from Parish 1953). Chart showing the proposed echo-search coverage of the North Sea. Dotted line indicates the proposed North Sea echosounder survey grid; solid line the approximate routes followed by some shipping lines and commercial trawl (A = Route worked by S/S St. Clair; B= Approximate route of British Arctic trawlers working from Humber ports).

Fig. 3: Survey area layouts and dates for all participating vessels in the 2000 international North Sea herring acoustic survey. Dotted area indicate area of survey overlap.

Details of the methods employed on the surveys are available in the "Manual for Herring Acoustic Surveys in ICES Divisions III, IV and VIa” (ICES, 2000a). This itself is largely based on the principles laid out in Simmonds et al. (1992). The methods are standardised as much as possible with all vessels using the same frequency (38 kHz) and type of scientific echosounder. This is calibrated by each vessel on every survey in the same manner and instrument settings are largely the same. Transects are spaced at a maximum of 15 nautical miles and vessels survey speed is 10-12 knots. Most vessels suspend operation at some point at night when the herring become unavailable to the echosounder due to vertical migration into surface waters, although the exact timing depends on the location of the survey.
The surveys differ in the particular strategy and method of allocating echo records to species (scrutiny). This is due to the different nature of fish concentrations encountered in the area, but in all cases echo traces are verified by regular "ground-truth” trawling with either a pelagic or demersal trawl. Analysis of the scrutinised acoustic data is then standardised according to the procedures in the manual with the same target strength conversion factors being applied (e.g. TS = 20 log L –71.2 for herring). The estimates are split up according to age and maturity using appropriate local age/maturity length keys derived from the trawl samples. Biomass is obtained by applying length weight relationships also derived from the trawl samples.
The results from the individual acoustic surveys are reported to the ICES Planning Group for Herring Surveys (e.g. ICES, 2000a) in a standard form, namely abundance in numbers at age, mean weight at age and total biomass. The individual surveys are then combined to produce a global estimate which takes into account any survey overlaps.
The global estimate of numbers at age and mean weight at age are then reported to the ICES Herring Assessment Working Group (ICES, 2000b) which use the data to tune the Integrated Catch at Age model (Patterson and Melvin, 1996) which ultimately produces the biomass estimates for the stock.

References

Au, W. (1993). The Sonar of Dolphins. New York, Springer Verlag.277 pp.

Bailey, M. C., Maravelias, C. D. and Simmonds, E. J. (1998). Changes in the distribution of autumn spawning herring (Cupea harengus L.) derived from annual acoustic surveys during the period 1984-1996. ICES Journal of Marine Science 55: 545-555.

Cushing, D. H. (1973). The detection of fish. Oxford, Pergamon Press.200 pp.

Fernandes, P. G., Gerlotto, F., Holliday, D. V., Nakken, O. and Simmonds, E. J. (2001). Acoustic applications in fisheries science: the ICES contribution. ICES Journal of Marine Science In press.

Foote, K. G. (1983). Linearity of fisheries acoustics, with additional theorems. Journal of the acoustical society of America 73: 1932-1940.

Forbes, S. T. and Nakken, O. (1972). Manual of methods for fisheries resource and appraisal. Part 2. The use of acoustic instruments for fish detection and abundance estimation. FAO Manual in Fisheries Science 5: 138.

Hewitt, R. P. and Demer, D. A. (1991). Krill abundance. Nature 353: 310.

ICES (1980). Report of the herring assessment working group for the area south of 62°N. ICES CM 1980/H:4 103 pp.

ICES (1989). Report of the planning group for acoustic surveys in sub-area IV and Division IIIa. ICES CM 1989/H:3 14 pp.

ICES (2000a). Report of the planning group for herring surveys. ICES CM 2000/G:02 106 pp.

ICES (2000b). Report of the herring assessment working group for the area south of 62°N. ICES CM 2000/ACFM:10 409 pp.

Kloser, R. J. (1996). Improved precision of acoustic surveys of benthopelagic fish by means of a deep-towed transducer. ICES Journal of Marine Science 53(2): 407-413.

MacLennan, D. N. and Simmonds, E. J. (1992). Fisheries acoustics. London, Chapman & Hall.325 pp.

Medwin, H. and Clay, C. S. (1998). Fundamentals of acoustical oceanography. San Diego, Academic Press.712 pp.

Mitson, R. B. (1983). Acoustic detection and estimation of fish near the seabed and surface. FAO Fisheries Report 300: 24-37.

Mitson, R. B. (1984). Fisheries Sonar. London, Fishing News Books.287 pp.

Ona, E. and Mitson, R. B. (1996). Acoustic sampling and signal processing near the seabed: the deadzone revisited. ICES Journal of Marine Science 53: 677-690.

Parrish, B. B. (1953). A proposal for the introduction of organised echo-search in North Sea herring investigations. ICES CM 1953/Pelagic Committee:30 pp.

Patterson, K. R. and Melvin, G. D. (1996). Integrated Catch at Age analysis Version 1.2. Scottish Fisheries Research Report 38.

Reid, D. G. (2000). Echo trace classification. ICES Co-operative Research Report 238: 115.

Simmonds, E. J., Williamson, N. J., Gerlotto, F. and Aglen, A. (1992). Acoustic survey design and analysis procedure: a comprehensive review of current practice. ICES Cooperative Research Report 187: 127.

Urick, R. J. (1983). Principles of underwater sound. New York, McGraw Hill.423 pp.

Walsh, S. J., Engås, A., Ferro, R. S. T., Fontaine, R. and Van Marlen, B. (2001). Improving fishing technology, to catch (or conserve) more fish: the evolution of the ICES fishing technology & fish behaviour group during the past century. ICES Journal of Marine Science. In press.

Entered/Updated by (Date)
Paul Fernandes (11 Jan 01)