Smith, S. L., B. H. Jones, L. P. Atkinson, and K. H. 

 Brink. 



1986. Zooplankton in the upwelling fronts off Pt. 

 Conception, California. In J. A. C. Nihoul (editor), 

 Marine interfaces ecohydrodynamics, p. 195-213. 

 Elsevier Oceanogr. Ser. , 42. 

 Sokal, R. R., and F. J. Rohlf. 



1981. Biometry. 2nd ed. W. H. Freeman, N.Y., 

 859 p. 

 Tranter, D. J., and P. E. Smith. 



1968. Filtration performance. In D. J. Tranter 

 (editor), Zooplankton sampling, p. 27-56. Monogr. 

 Oceanogr. Methodol. 2. UNESCO Press, Paris. 

 Venrick, E. L., and T. L. Hayward. 



1984. Determining chlorophyll on the 1984 CalCOFI 

 surveys. Calif. Coop. Oceanic Fish. Invest. Rep. 

 25:74-79. 

 Wiebe, P. H., S. Boyd, and J. L. Cox. 



1975. Relationships between zooplankton displace- 

 ment volume, wet weight, dry weight, and carbon. 

 Fish. Bull., U.S. 73:777-786. 



M. D. Ohman 

 J. R. Wilkinson 



Scripps Institution of Oceanography 



MLRG A-027 



La Jolla. CA 92093 



T\ina Larvae Abundance: Comparative 



Estimates from Concurrent Japanese and 



Australian Sampling Programs 



When estimating the absolute abundance of or- 

 ganisms, the accuracy and bias of sampling 

 methods should be assessed (Andrew and Map- 

 stone 1987). In ichthyoplankton sampling the ab- 

 solute abundance of organisms will probably 

 never be known; the characteristics of accuracy 

 and bias in different sampling methods can only 

 be inferred by concomitant samphng of the same 

 population. The Fishery Agency of Japan Far 

 Seas Fisheries Research Laboratory (FSFRL) 

 has used a 2 m ringjieLtojample^ichthyQplank- 

 ton for many years. It has been the principal tool 

 for sampling tuna larvae, particularly southern 

 bluefin tuna, in the eastern Indian Ocean (Yabe 

 et al. 1966; Ueyanagi 1969; Yonemori and Morita 

 1978; Yukinawa and Miyabe 1984; Yukinawa and 

 Koido 1985). The net routinely samples large 

 volumes of water (approximately 5,000 m'^ in a 

 30-min oblique tow), yet catches of tuna larvae 

 on these surveys are generally low. These low 

 catches may reflect a naturally low abundance of 

 tuna larvae, a contention supported by previous 

 studies (Wade 1951; Strasburg 1960; Klawe 1963; 



Richards 1969; Richards and Simmons 1971; 

 Conand and Richards 1982). In this paper we 

 compare catches of tuna larvae by traditional 

 Japanese methods with those developed by 

 CSIRO Division of Fisheries for quantitative 

 surveys. A series of simultaneous tows were 

 made" by the CSIRO, FRV Soela and the 

 FSFRL, FRV Shoyo Mnrn on the southern 

 bluefin tuna spawning gi-ounds in the east Indian 

 Ocean in January 1987. 



Methods 



Two identical 2 m ring nets were deployed 

 concurrently by the FSFRL (Fig. 1, Table 1) in 

 surface and obhque tows. For the oblique tow, a 

 predetermined length of warp (approximately 

 130 m) was rapidly paid out from the stern so 

 that the net reached a depth of 30 m (approxi- 

 mately 4-10 minutes). The warp was then re- 

 trieved at a fixed rate until the net reached the 

 surface (approximately 21-26 minutes). The tow 

 profile actually achieved was determined after 

 the tow from traces made by the depth distance 

 recorder. The 20-min surface tow was deployed 

 close to the hull on the starboard side, amid- 

 ships, with approximately 7/8 of the net below 

 the surface, fishing a depth range of 0-1.75 m. 



Two identical 70 cm ring nets were deployed 

 concurrently by CSIRO (Fig. 1, Table 1) in sur- 

 face and obhque tows. The oblique tow fished 

 from the surface to the thermocline (the thermo- 

 cline during the experiment was at approxi- 

 mately 32 m) to cover the full known depth range 

 of the tuna larvae (CSIRO, unpubl. data). An 

 operator, guided by real-time depth information 

 from a sensor on the net, produced a V-shaped 

 tow profile, with a descent time of approxi- 

 mately 8 minutes and an ascent of 12 minutes. 

 The surface tow was deployed for 10 minutes, 

 concurrent with the oblique tow, from a boom on 

 the port side amidships, clear of the wake of the 

 vessel. It was towed approximately 0.5 m under 

 the surface, oscillating between about and 2 m 

 due to the roll of the vessel in the 0.5 m swell. 



The volume of water filtered for each net was 

 calculated in the following ways. The volumes 

 filtered by both surface and oblique tows with 

 the 70 cm net were calculated from the distance 

 travelled, measured by calibrated flowmeter 

 readings inside the net, and the mouth area of 

 the net. Volumes filtered by oblique tows with 

 the 2 m net were calculated from the distance 

 travelled (determined from the depth distance 

 recorder behind the net) and the mouth area of 



Manuscript accepted April 1989. 

 Fishery Bulletin. U.S. 87:976-981. 1989. 



976 



