580 



Fishery Bulletin 105(4) 



rates of seabirds. From the point of view of the EB 

 method, this lack of correlation indicates that the Pols- 

 son catch rate, A,, was not a characteristic of individual 

 fishing vessels, but is largely due to chance in any one 

 year. Vessels 28 and 36 did not fish in 2003 and were 

 excluded in the comparison. More detailed studies with 

 current data should be carried out to determine if this 

 is a valid conclusion. 



As in 2002, the EB estimates for 1997 did not improve 

 on stratum ML estimates for estimating seabird bycatch 

 rates per vessel. However, ML and EB estimates of in- 

 dividual vessel bycatch rates in 1997 were found to be 

 significantly positively correlated with observed bycatch 

 rates for 1998. In this era of high bycatch rates of sea- 

 birds, bycatch rates were found to be more dependent 

 on the practices of individual vessels. 



