814 



Fishery Bulletin 92(4). 1994 



with seasonal migrations within a closed system. 

 Other movements occur against a backdrop which is 

 dominated by regular seasonal movements. 



Movements of Pacific cod may be better understood 

 in the context of the general life history, population 

 dynamics and physical environment requirements of 

 this species. Bakkala (1984) examined distribution 

 patterns based on an analysis of research survey and 

 commercial fishery catch per unit of effort and size 

 composition data. He described a gradual shift over 

 the southeast shelf with time, corresponding to a 

 progression in cohort ages and the influence of 

 year-class abundance. A tendency towards the off- 

 shore environment was noted from coastal waters to 

 the outer shelf and slope edge. This was based on an 

 areal transition stemming from ontogenetic develop- 

 ment in younger age ( 1-3 yr) to older age (4+ yr) groups. 

 Further, during years of higher than average abun- 

 dance, the population range was much more extensive 

 than that in years of low abundance (Bakkala, 1984). 



From Russian trawl surveys, Stepanenko 5 de- 

 scribed winter concentrations along the upper slope 

 at depths between 400 and 545 m. Prespawning and 

 spawning aggregations were consistently found 

 northwest of Unimak Island, in the Pribilof Islands 

 sector, and along the northern slope on either side of 

 the U.S. -Russia Convention Line. The most signifi- 

 cant spawning aggregations occurred in the vicinity 

 of Unimak Pass along the outer shelf edge. 



The literature indicates that preferred water tem- 

 peratures (0 to 10°C) are the primary factor for de- 

 termining centers of Pacific cod abundance. Towards 

 its southern range off British Columbia, Ketchen 

 (1961) reported highest catch rates at bottom tem- 

 peratures of between 6 and 9°C. Off Russia, Moiseev 

 (1953) noted that spawning Asian Pacific cod pre- 

 ferred 80-290 m depths and water temperatures be- 

 tween and 2 to 3°C; optimal summer temperatures 

 were between 0.2 and 4.5°C. Hirschberger and Smith 

 ( 1983 ) reported water temperatures around 5.4°C for 

 spawning Gulf of Alaska Pacific cod at 150-250 m. 



In the eastern Bering Sea, high winter concentra- 

 tions of Pacific cod coincide with warmer water ( mean 

 4.0°C) found year-round in depths off Unimak Pass 

 and the upper slope ( Kihara, 1982, a and b; Bakkala, 

 1984). Bottom temperatures on the shelf drop from 

 the 0.2 to 4.5°C range in summer to below 0°C in 

 winter (Schumacher and Reed, 1983). Thus at the 

 high latitudes of the Bering Sea, the stimulus for 



5 Stepanenko, M. A. 1989. Condition of stocks, interannual vari- 

 ability of catch per unit of effort and fishing of cod in the east- 

 ern part of the Bering Sea. Pacific Research Institute of Fisher- 

 ies and Oceanography (TINRO), Vladivostok, USSR. Document 

 submitted to the US-USSR bilateral meetings, November 1989, 

 35 p. Available: Alaska Fish. Sci. Cent., NOAA, NMFS, 7600 

 Sand Point Way NE., Bin C15700, Seattle, WA 98115-0070. 



offshore migration appears to be avoidance of the 

 intense cooling of inshore waters that accompany 

 advancing ice formation from the Bering Strait in 

 favor of warmer temperatures at depth. The spring 

 feeding migration, shoreward, is most likely timed 

 to the warming of the coastal shelf environment and 

 a return to summer norms (Bakkala, 1984). 



Interestingly, at lower latitudes, seasonal migra- 

 tions are reversed. At the southernmost edge of its 

 range, off Korea, Japan, and in Puget Sound, Wash- 

 ington (Karp, 1982; Mishima, 1984; Zhang, 1984), 

 Pacific cod migrate to deep offshore waters during 

 summer months to avoid excessively heated (>10°C 

 coastal waters. A returning inshore spawning migra- 

 tion occurs each winter. 



Moiseev (1953) noted very limited along-shore 

 migrations in Russian waters. However, active sea- 

 sonal migrations between coastal shallows and off- 

 shore depths perpendicular to the shoreline (mainly 

 in response to inshore and offshore temperature 

 shifts) were found. He further hypothesized that the 

 potential for stock intermingling was reduced because 

 of limited along-shore movement. Winter offshore move- 

 ments were observed throughout the northern range 

 of Pacific cod. Local abundance centers were always in 

 the direction of the preferred temperature regime in 

 response to the pronounced cooling of their onsheff en- 

 vironment. Westrheim ( 1984) noted that Pacific cod off 

 Vancouver Island, British Columbia, exhibited the same 

 bathymetric seasonal movements as Pacific cod in 

 Alaska but very limited along-shore movements. 



In all of the cases described above, seasonal mi- 

 grations of Pacific cod appear to be triggered by the 

 desire to avoid temperature extremes that accom- 

 pany the changing seasons. In the eastern Bering 

 Sea, movements represent necessarily long-range 

 migrations across the Bering Sea shelf which on av- 

 erage is 300 nmi wide. For southern coastal stocks, 

 the same result can be achieved with much shorter 

 offshore migrations to depth. It is likely that inter- 

 regional along-shore migrations seldom are found 

 because they are unnecessary for achieving the pre- 

 ferred temperature regime (Moiseev, 1952). 



Rose ( 1993) found a similar pattern in Atlantic cod, 

 Gadus morhua, based on hydroacoustic surveys. He 

 attributed seasonal movement to springtime feeding 

 migrations, which shifted Atlantic cod from offshore 

 winter spawning grounds shoreward. Migration path- 

 ways were facilitated by stable bottom temperature 

 regimes (2— 3°C) associated with trenches on the north- 

 eastern Newfoundland shelf. 



How large-scale stock structure is affected by mi- 

 grations motivated by preferred temperatures is 

 unclear. Grant et al. (1987) screened Pacific cod ge- 

 netic samples from throughout their range. Two ge- 



