Shimada and Kimura: Seasonal movements of Gadus macrocephalus 



813 



Pacific side of Akutan Pass and Unimak Pass in the 

 Aleutian Islands. Additionally, two fish tagged in the 

 major spawning area (Area 3) were recaptured to the 

 west in Seguam Pass within 250 days. In a striking 

 occurrence of immigration to the Bering Sea, a pair 

 of Pacific cod (65 cm fork length) tagged in Tanaga 

 Pass near Adak Island were recaptured on the outer 

 northwest shelf (above 57°N) after 3 and 5 years at 

 liberty (Fig. 8). 



Although substantial numbers of Pacific cod were 

 tagged along the Aleutian Islands west of 170°W, in- 

 cluding to about 174°E (Fig. 1), few recoveries have 

 been made. These releases came from a single 1986 

 summer trawl survey and tagged fish were in poor 

 condition because commercial foreign fishing opera- 

 tions were employed. 



Discussion 



Our analysis identified a seasonal circuit that we 

 attribute to annual migrations for spawning and feed- 

 ing; it also provided preliminary indications of emi- 

 gration from the eastern Bering Sea. The former is 

 described in terms of three eastern Bering Sea ar- 

 eas. The latter ties together more expansive distances 

 as defined by regional geography or fishery manage- 

 ment boundaries, or both (OCSEAP, 1987). Although 



the majority of tagged Pacific cod exhibited the sea- 

 sonal character of short-term cross-shelf movements, 

 a small number of tagged individuals provided em- 

 pirical evidence for much longer transits. 



We recognize that emigration and immigration 

 probably occur with respect to the main study area 

 (i.e. Areas 1-3). However, the locations of tag releases 

 and numbers of tag recoveries received to date make 

 it difficult either to quantify the amount of emigra- 

 tion from the eastern Bering Sea, or to conclude with 

 certainty that return immigration to the eastern 

 Bering Sea is significant. At this time, we believe 

 there is some eastern Bering Sea immigration from 

 the surrounding regions. However, considerable un- 

 certainty exists because so few Pacific cod were 

 tagged outside the study region, most importantly 

 in the central and eastern Gulf of Alaska (Fig. 1). 

 Also, we have some evidence that the western Aleu- 

 tian Islands stock(s) may be fairly independent of 

 the eastern Bering Sea, but this evidence is far from 

 conclusive. 



Despite these conjectures, statistics such as "dis- 

 tance traveled" and "rate traveled" versus "time at 

 liberty" (Fig. 9) generally support our seasonal move- 

 ment model. Observed "distance traveled" already is 

 maximized within the first year of freedom. Simi- 

 larly, observed "rate traveled" is maximized within 

 the first year at liberty. This behavior is consistent 



E 

 ■o 



> 



CO 



1— 



Q> 

 O 



c 



03 



a 



o 

 o 



o 

 o 

 ro 



o 

 o 



o 



500 1000 1500 

 Time at liberty (days) 



2000 



co 



I 



T3 



a> 

 S3 



m 



 



1 1 



1500 2000 



500 1000 



Time at liberty (days) 



Figure 9 



Scatter plots showing (left) the relationship between distance traveled (in nautical miles) and time at liberty and 

 (right) the relationship between rate of travel and time at liberty for Pacific cod. 



