Doyle et al Neustonic ichthyoplankton in the western Gulf of Alaska 



251 



these larger specimens undertake a nocturnal mi- 

 gration into the neuston as has been indicated by 

 data collected off the U.S. west coast (Doyle, 1992). 

 The length-frequency distributions documented for 

 A. hexapterus larvae here, however, suggest that 

 during spring in the western Gulf of Alaska, the well- 

 developed larvae and early juveniles (>20 mm SL) 

 almost exclusively occupied the neuston. Such speci- 

 mens were rare in the bongo net samples where the 

 predominant larval size range was 5-15 mm SL. The 

 scarcity of the large A. hexapterus larvae in the day- 

 time neuston samples in this instance could be at- 

 tributed to light-enhanced sampler avoidance. 



Brodeur and Rugen (1994) also found that 

 Bathymaster spp. larvae (4-7 mm SL) were deepest 

 in the water column at night in the western Gulf of 

 Alaska and suggest a diel migration pattern of noc- 

 turnal descent similar to A. hexapterus. A similar 

 pattern of downward migration at night has been 

 observed for Bathymaster spp. larvae in the Bering 

 Sea ( Walline 6 ). Their absence from daytime neuston 

 samples during the present study seemed to contra- 

 dict such a pattern of vertical migration. Whereas 

 most of the young larvae may follow the above pat- 

 tern, our data also suggest a facultative association 

 with the neuston by some of these larvae and a night- 

 time occupation of the neuston as a result of migra- 

 tion upward to the surface. This seems feasible as it 

 is apparent from observations by Kendall and Dunn 

 (1985) and Rugen 3 in the Gulf of Alaska that 

 Bathymaster spp. larvae become more neustonic with 

 development. Most Bathymaster spp. larvae taken 

 here, both in neuston net and bongo net samples, 

 were <10 mm SL and would not be able to avoid the 

 sampler as did the A. hexapterus larvae found in the 

 neuston during the present study. The facultative 

 nocturnal association with the neuston proposed here 

 for Mallotus villosus larvae does not contradict what 

 is already known concerning vertical distribution 

 patterns for osmerids in the Gulf of Alaska. 

 Haldorson et al. ( 1993) recorded that osmerid larvae 

 in Auke Bay apparently spend most of their time in 

 the mixed layer, rising to the surface at night and 

 returning to relatively shallow depths during the day. 



Recent investigations on the interaction between 

 the early life history stages of T. chaleogramma and 

 the oceanographic environment in the western Gulf 

 of Alaska indicate that prevailing southwesterly cur- 

 rents transport larvae from the Shelikof Strait re- 

 gion to nursery grounds along the Alaska Peninsula 



(Kendall et al., 1987; Kim and Kendall, 1989; 

 Hinckley et al., 1991; Schumacher and Kendall, 

 1991). Although the southwesterly flowing Alaska 

 Coastal Current bifurcates southwest of Kodiak Is- 

 land, most of this water remains on the shelf, thus 

 potentially retaining the majority of fish larvae in 

 the coastal region. Physical features such as plumes 

 and eddies also serve to retain larvae on the conti- 

 nental shelf and transport them southwestward 

 along the Alaska Peninsula (Vastano et al., 1992). 



Given these current patterns, the distribution pat- 

 terns observed for most taxa of fish larvae in the 

 neuston during this study suggest that springtime 

 spawning and emergence of larvae into the plank- 

 ton (and subsequently the neuston) took place mainly 

 around Kodiak Island (except along the seaward side) 

 and along the Alaska Peninsula to the southwest. A 

 high concentration of larvae over the shelf from 

 Kodiak Island to the Shumagin Islands was the pre- 

 dominant pattern for most species. Despite their oc- 

 currence in the neuston, these larvae were likely re- 

 tained over the shelf and in the coastal zone by the 

 prevailing currents. In contrast, the more offshore 

 distribution patterns observed for A. fimbria, P. 

 monopterygius, and H. hemilepidotus indicate that 

 a significant proportion of these larvae may have 

 been entrained in the Alaskan Stream over the slope 

 and in deep water. 



Analysis of multispecies spatial patterns using 

 recurrent group analysis and numerical classifica- 

 tion did not reveal the existence of more than one 

 neustonic assemblage offish larvae in the study area. 

 A unique and comparable assemblage of neustonic 

 fish larvae has also been identified off the U.S. west 

 coast and its geographical distribution is essentially 

 confined to shelf and slope waters off Washington, 

 Oregon, and northern California (Doyle, 1992; 

 Doyle 7 ). Apart from perhaps P. monopterygius lar- 

 vae, which are known to occur throughout the Gulf 

 of Alaska (Gorbunova, 1962), and to a lesser extent 

 A. fimbria and H. hemilepidotus, members of the 

 neustonic assemblage offish larvae in the western Gulf 

 of Alaska are likely to be scarce in the oceanic zone. 



It has been postulated that the primary advantage 

 of a neustonic existence as an early life history strat- 

 egy for certain species of marine fish is the enhanced 

 trophic conditions that prevail in this biotope ( Moser, 

 1981; Tully and O'Ceidigh, 1989; Doyle, 1992). The 

 suitability of the neuston as a feeding ground for lar- 

 vae is, however, dependent on the ability of larvae to 



6 Walline, P. D. 1981. Hatching dates of walleye pollock I Theragra 

 chaleogramma) and vertical distribution of ichthyoplankton 

 from the eastern Bering Sea, June-July 1979. U.S. Dep. Commer., 

 NOAA, Natl. Mar. Fish. Serv., Alaska Fish. Sci. Cent., 7600 Sand 

 Point Way NE, Seattle, WA 98115. Proc. Rep. 81-05, 22 p. 



Doyle, M. J. 1992. Patterns in distribution and abundance of 

 ichthyoplankton off Washington, Oregon, and northern Cali- 

 fornia (1980 to 1987). U.S. Dep. Commer., NOAA, Natl. Mar. 

 Fish. Serv., Alaska Fish. Sci. Cent., 7600 Sand Point Way NE, 

 Seattle, WA, 98115. Proc. Rep. 92-14, 344 p. 



