absolute size, length composition, and overall sex ratio. During 

 both surveys, juvenile ( <2 yr) distributions were not geographic- 

 ally separated from adults (5-10 yr), with highest abundances of 

 young individuals occurring along the outer continental shelf in 

 subareas 2 (1975 and 1976) and 3N (1975 only). 



Pacific cod. — Although Pacific cod ranged broadly over near- 

 ly the entire eastern Bering Sea continental shelf, during both the 

 1975 and 1976 surveys most of the population was distributed 

 along the outer shelf and slope (Fig. 87). Occurrences in inner and 

 central regions were represented by only low densities, and in 

 general, seasonal differences in these areas were unclear. Along 

 the outer shelf and slope, however, Pacific cod abundances were 

 higher and showed a more continuous pattern of high densities 

 during April- June 1976 than during August-October 1975. 



Like walleye pollock, attempts to assess the Bering Sea popula- 

 tion of Pacific cod included formidable sampling problems due to 

 unknown semidemersal behavior, potential trawl avoidance, 

 dense shoal formation, and potentially large effects due to in- 

 migration and out-migration. Although midocean pelagic popula- 

 tions have not been reported, substantial migratory exchange may 

 occur between populations over the continental slope (depths 

 180-1,000 m) and populations on the outer shelf (depths 

 100-180 m). 



A feature common to both the 1975 and 1976 survey results was 

 that the geographical distributions of size and age classes were 

 strongly related to bottom depth. Juveniles (0-group and 1-yr- 

 olds) apparently develop in the shallow central and inner shelf 

 regions, then progressively move to deeper water. Populations in 

 deep water were mainly composed of large, older individuals in 

 contrast to populations in shallower waters. However, the max- 

 imum age observed even in deepwater populations was only 6 yr. 



Alaska plaice. — As observed for yellowfin sole, the distribu- 

 tional patterns of Alaska plaice differed markedly between the 

 1975 and 1976 surveys (Fig. 88, 89). During August-October 1975, 

 Alaska plaice ranged broadly over the central and inner shelf, 

 with two principal centers of abundance. During April-June 1976, 

 a relatively high-density front of migrating individuals appeared 

 to move from a winter retreat in deep water, approaching a large- 

 scale pattern similar to the late-summer distribution observed dur- 

 ing 1975. 



Although sampling problems resulting from the population's 

 movements in 1976 prevented meaningful comparisons of abun- 

 dance estimates, there were similarities in size and age structure, 

 and overall sex ratio between the two surveys. During both 1975 

 and 1976, juvenile Alaska plaice were most abundant in the 

 northern, central region of the continental shelf (subareas 4S and 

 4N). 



cause the vertical range of all three species may extend to bottom 

 depths of 500 to 1,000 m — beyond normal trawl sampling capa- 

 bilities, then in-migration and out-migration effects were poten- 

 tially large due to seasonal movements between bathymetric 

 zones, particularly in regions with steep bottom slope; 2) trawl 

 avoidance may have been substantial by the large, strong- 

 swimming adults; and 3) tagging studies have indicated that out- 

 migration to, and in-migration from, populations in other regions 

 may be significant (Dunlop et al. 1964; International Pacific 

 Halibut Commision 1973). Additionally, relationships between 

 the southeastern Bering Sea population of arrowtooth flounder, 

 Atheresthes stomias, and reported northwestern Bering Sea pop- 

 ulation of the congeneric Asian arrowtooth flounder, A. ever- 

 manni, are unclear (Wilimovsky et al. 1967). 



During August-October 1975, Greenland turbot were broadly 

 distributed over central and outer regions of the continental shelf 

 at low densities, and one large northern center of abundance rep- 

 resented over one-half the total apparent population (Fig. 90). 

 During April- June 1976, population densities were high along the 

 southwest shelf edge (subarea 2 Slope) in deep water (Table 66), 

 and extension onto the shelf was restricted. 



In general, shelf populations of Greenland turbot during both 

 surveys were primarily juveniles (<4 yr of age). One-year-old-in- 

 dividuals were most abundant in subareas 4S and 4N (1975 and 

 1976) and subarea 3 (1976 only). Deepwater populations (par- 

 ticularly in subarea 2 in 1975 and subarea 2 Slope in 1976) were 

 composed of large, old (3-16 yr) individuals. 



In contrast to Greenland turbot, arrowtooth flounder distribu- 

 tions varied less between the two surveys and northern popula- 

 tions were small (Fig. 91). During both surveys, most arrowtooth 

 flounder occurred along the outer shelf and slope between St. 

 George and Unimak Islands in subareas 2 (1975 and 1976) and 2 

 Slope (1976 only). During 1975, several high-density locations oc- 

 curred at intervals along the outer continental shelf. In 1976, 

 abundance was high only along the shelf edge and slope (Table 

 67). Juveniles (ages 1-3 yr) were most abundant in subarea 2 dur- 

 ing both years. 



Of the three deepwater flounder species, Pacific halibut showed 

 the most extreme differences in geographical distributions be- 

 tween the two surveys (Fig- 92). During August -October 1975, the 

 apparent population was broadly distributed in Bristol Bay with 

 scattered occurrences on the central and outer shelf. During 

 April- June 1976, Pacific halibut were distributed primarily along 

 the southwest continental shelf between St. George and Unimak 

 Islands and were abundant at slope depths (Table 68). Although 

 large individuals may have been underestimated due to avoidance, 

 trawl catches during both surveys indicated that shelf populations 

 were mainly juveniles (20-50 cm FL). 



Other flounders. — The three pleuronectid species — Greenland 

 turbot, arrowtooth flounder, and Pacific halibut — provided simi- 

 lar population assessment problems, and survey results for the 

 three species showed many parallels. All three species have exten- 

 sive geographical ranges, occurring along the entire Pacific rim 

 from southern California to the Kamchatka Peninsula and along 

 all continental shelf regions of the Bering Sea (Hart 1973). In ad- 

 dition, the regional populations of all three species appear to 

 undergo regular seasonal migrations to the upper continental 

 slope or shelf in summer and then return to deep water during 

 winter where spawning occurs (Novikov 1964; Musienko 1970; 

 Shuntov 1970). 



Population assessment problems included the following: 1) Be- 



Table 66. — Apparent bathymetric distribution of Greenland 

 turbot abundance within the 1976 Bering Sea slope subarea. 1 







Mean CPUE : 



Bottom depths (m) 



No. of samples 



(kg/km) 



200-249 



10 



5.2 



(0.0-18.4) 



250-299 



8 



10.6 



(0.3-42.6) 



300-349 



9 



10.5 



(0.3-29.9) 



350-399 



7 



32.9 



(5.0-98.4) 



400-449 



7 



24.0 



(0.0-105.2) 



450-460 



3 



66.2 



(31.8-133.9) 



'See Figure 3. 



: Mean catch per unit fishing effort, with range in paren- 

 theses. 



121 



