134 



Fishery Bulletin 90(1). 1992 



Horizontal distribution In the Funka Bay region, 

 walleye pollock larvae are generally concentrated 

 inside the bay from late January through early April 

 (Nakatani 1988, Nakatani and Maeda 1989). Their 

 abundance decreases during this time from >5000 

 larvae/m- in the area of maximum concentration in 

 late January to 200-400 larvae/m^ in early April. In 

 many cases, surveys have disclosed more than one area 

 of abundance within the bay. Their occurrence general- 

 ly overlaps that of the Oyashio Water. For example, 

 in 1980 the Tsugaru Warm Water remained in the bay 

 longer than usual, and the Oyashio Water did not enter 

 the bay until mid-March; before then, the larvae were 

 concentrated at the mouth of the bay. It is possible that 

 larvae entering the bay before the invasion of the 

 Oyashio Water would experience low survival because 

 of inadequate prey production. 



In Shelikof Strait, most larvae are concentrated in 

 one large patch that can be followed as it drifts to the 

 southwest with the prevailing currents from April 

 through May (Kendall et al. 1987). The velocity of drift 

 may vary interannually and depend on weather pat- 

 terns in the area as well as the strength of the ACC. 

 In some years, it appears that most of the larvae drift 

 out of the strait within 2-4 weeks after hatching, but 

 in other years they remain for several more weeks 

 because of the influence of nearshore eddies (Incze 

 et al. 1989). There is considerable cross-strait shear in 

 the current, so the drift of larvae is influenced by where 

 they reach the surface layer from their deep incuba- 

 tion area (Kim and Kendall 1989). Larval abundances 

 as high as lO.OOO/m- were observed in the patch in 

 late April 1981, and by late May abundances of 2400/ 

 m^ were present (Bates and Clark 1983). 



Feeding Copepod nauplii, which were not identified 

 to species, are the major prey item of first-feeding 

 walleye pollock larvae (Kamba 1977, Kendall et al. 

 1987, Nakatani and Maeda 1983). Copepodids are the 

 most important prey item in the diet of 1 1 mm larvae 

 in Shelikof Strait and 8 mm larvae in Funka Bay. 

 Copepod eggs were more prevalent in guts of larvae 

 in Funka Bay than in Shelikof Strait (Nakatani and 

 Maeda 1983, Kendall et al. 1987). Their digestibility and 

 nutritional value for walleye pollock larvae are un- 

 known. Pseudocalanus spp. was the most abundant 

 copepod taxon in the water column in Shelikof Strait 

 and Funka Bay when larvae were present (Kendall 

 et al. 1987, Nakatani 1988). The nauplii in the guts of 

 small larvae were probably mostly Pseudocalanus spp. 

 and Oithona spp., and most of the copepodids in larger 

 larvae were Pseudocalanus spp. Copepodids oi Pseudo- 

 calanus minutus and Oithona similis were most 

 abundant in larger larvae up to 30 mm in Funka Bay 

 (Nakatani and Maeda 1983). The maximum prey size 



increases with growth of the larvae, but the minimum 

 size remains fairly constant through fish up to about 

 73 mm (Kamba 1977). 



Based on laboratory and field studies, naupliar abun- 

 dances of about 10 per liter seem to be required to sup- 

 port growth of small (<8mm) walleye pollock larvae 

 (Paul 1983, Dagg et al. 1984). Prey densities above this 

 threshold have been observed associated with the larval 

 patch in Shelikof Strait before and during a storm 

 (Incze et al. 1990). Naupliar abundances below this 

 threshold were seen in Funka Bay throughout most of 

 the larval period in 1987, but they were above 10 per 

 liter in several other years (Nakatani and Maeda 1989). 

 However, naupliar densities were probably underesti- 

 mated, since they were collected on 100/jm sieves. 

 AvaUabOity of smaller nauplii as larval food will require 

 further observations. 



Age and growth Daily growrth increments on oto- 

 liths have been used to determine the age of larvae and 

 early juveniles from both Shelikof Strait and Funka 

 Bay. Based on a series of 109 larvae (6.0-14.6 mm SL) 

 collected in Shelikof Strait in May 1983, the linear 

 growth equation SL = 4.29mm -t- 0.21 d (r- 0.75), 

 where d = age in days, was fit (Kendall et al. 1987). 

 Growth based on 357 larvae and early juveniles 3.9- 

 30.0mm SL from the Shelikof spawning collected May 

 through July 1987 fit a Laird-Gompertz function: SL 

 at age t = 4.505 (e''-854(i-e-"'*'"))^ where t = days after 

 hatch (Yoklavich and Bailey 1989). The growth of 

 larvae and juveniles from Funka Bay fit the function: 

 TL = 121.5/(1 -He-o-026(t-i24.5ii))^ with TL in mm 



(Nishimura and Yamada 1984). Thus larvae 50 days old 

 from Funka Bay were about 14.0mm SL (see footnote) 

 while those from Shelikof Strait would range from 14.8 

 mm SL (Kendall et al. 1987) to 18.7mm SL (Yoklavich 

 and Bailey 1989) (Fig. 4). 



Larval population length-frequency distributions de- 

 pend on time of spawning, mortality of larvae, growth 

 of larvae, and sampling bias. Except for sampling bias, 

 these factors represent population processes occurring 

 to the annual cohort of larvae. In Funka Bay, even 

 though spawning takes place over a protracted period, 

 larval survival appears low except during periods when 

 adequate food is present. Mortality due to starvation 

 is high for larvae that hatch before the spring increase 

 of nauplii in Funka Bay (Nakatani and Maeda 1989, 

 Nakatani 1991). Thus variations in size of larvae may 

 depend more on differences in the birth dates of sur- 

 viving larvae than on differences in growth rates. 



In Shelikof Strait, spawning peaked during the first 

 week of April in several years. By the end of April 

 1981, most larvae were about 4.8mm. By the third 

 week in May 1981, they were mostly 7-8mm (Dunn 

 et al. 1984), as they were in 1982 (Kendall et al. 1987). 



