Boje: Intermingling and seasonal migrations of Reinhardtius hippoglossoides 



419 



the Qord (Simonsen and Boje^). The catch-adjusted recap- 

 ture data were categorized by stratifying the fjords into 

 five rectangles of equal area, ranging from the mouth of 

 the fjord to the inner part of the fjord. A chi-square test on 

 the real recapture distribution compared with a random- 

 walk distribution based on pooled release data, proved 

 significant (P<0.0001). indicating that Greenland halibut 

 do not migrate randomly in the fjord system, but tend to 

 aggregate at specific localities at certain times of the year. 

 Pooling the data by ice-cover period (1^^ and 2""' quarters) 

 and open-water period (3'''^ and 4'*' quarters) and weighing 

 the distribution within each quarter equally also resulted 

 in significantly different distributions between the real 

 data and the random-walk data. The real pooled data, 

 shown by quarters (Fig. 4). demonstrated that Greenland 

 hahbut display a dispersed distribution during the ice- 

 cover period (P' and 2"'' quarters) but concentrate in the 

 inner parts of the Qords during the open-water period (3'^'^ 

 and 4"^ quarters). The steep gradient from the 4"^ to the 

 1st quarter may be due to the fishing practices of the com- 

 mercial fisheries, which are virtually dormant between 

 November and February. Thus fish recaptured from the 

 fishery for the two quarters mainly represent only the 

 months of October and March, respectively . 



' Simonsen, C. S.. and J. Boje. 2000. An assessment of the 

 Greenland halibut stock component in NAFO Division lA 

 inshore. NAFO ( Northwest Atlantic Fisheries Organziaton) SCR 

 Doc. 00/47, ser N4278, 37 p. NAFO Secretariat, 2 Morris Drive, 

 P. O. Box 638, P. O. Box 638, Dartmouth, Nova Scotia, Canada 

 B2Y 3Y9. 



Discussion 



Sampling by hook-and-line gear does not reflect the true 

 abundance of Greenland halibut in prespawning and 

 spawning conditions. But, because the majority of the tag- 

 ging took place outside the spawning period (winter and 

 early spring), the sampling gear should not have biased 

 any recovery patterns. Tagging in the southwest Green- 

 land fjords took place in January, but the recovery pat- 

 tern for those areas was confirmed by tagging data from 

 other seasons of the year (Riget and Boje, 1989). There- 

 fore, I assumed that any tagged specimens represent feed- 

 ing fish, not spawning groups of the population. 



Although Greenland halibut demonstrated long-dis- 

 tance displacements of about 2500 km in a two-year 

 period, most recoveries indicated rather limited move- 

 ment. The majority of tagged and recaptured Greenland 

 halibut in this study support the current understanding 

 of stock relationships among Greenland halibut popula- 

 tions (Bowering and Brodie, 1995; Boje et al., 1997; Vis 

 et al., 1997). However, several cases of single recoveries 

 suggest links between halibut populations not considered 

 before. One fish recorded off Newfoundland had been re- 

 leased in Godthaab Fjord two years earlier, yet currently 

 it is believed that Greenland halibut from Godthaab Fjord 

 and other fjords of southwestern Greenland originate 

 from the Icelandic spawning stock (Smidt, 1969; Riget et 

 al., 1989; Boje, 1993). Released Greenland halibut from 

 Baffin Bay yielded single recoveries in the northwestern 

 Greenland fjord of Torsukkattak and in Denmark Strait. 

 The latter is a migration distance of about 2500 km. Also, 

 one Greenland halibut released in eastern Greenland at 



