Van Eeckhaute et al Movements of Me/anogrammus oeglefinus determined from a population model 



663 



lll-18?m 



>183m, 



Si rata 



Figure 2 



Strata for DFO and NMFS bottom trawl surveys with depth ranges. Several NMFS strata are 

 bisected by the Canada-U.S. boundary line. Some strata for both countries are bisected by the 

 5Zj and 5Zm unit area boundaries. 



The DFO surveys have been conducted by the RV 

 Alfred Needier and for one year, by its sister ship, 

 the RV Wilfred Templeman , with a Western IIA trawl. 

 The NMFS surveys have been conducted by the RV 

 Albatross IV and the RV Delaware II with a BMV 

 door until 1984 and a polyvalent door subsequently. 

 A further gear alteration occurred with the replace- 

 ment of the standard Yankee 36 trawl by a modified 

 Yankee 41 trawl during the spring surveys in 1973- 

 81 (Hayes and BuxtonM. Conversion factors to ac- 

 count for these differences were not considered in 

 this study because the analysis involved only within- 

 year comparisons. In the autumn of 1985, 1986, and 

 1988, both NMFS vessels made tows in the 5Zjm area 

 but there was only one instance out of the three years, 

 1988, which would have resulted in slightly differ- 

 ent ratios. The boat conversion factor was therefore 

 not applied in our study. 



Abundance of haddock on each side of the ICJ line 

 was obtained by summing the estimates of abun- 

 dance for the strata in the respective jurisdictions. 

 This method can be applied directly to the Canadian 

 side for the DFO survey results because the strata 

 boundaries incorporate the ICJ line; however, the 

 strata on the U.S. side, 5Z3 and 5Z4, are bisected by 

 the 5Zj,m unit area lines (Fig. 2). Several strata in 

 the NMFS surveys are bisected by the ICJ line or 

 the 5Zj,m unit area lines, or by both (Fig. 2). We re- 

 fer to the parts of the bisected strata as strata sec- 

 tions. Because the tow locations were selected at ran- 

 dom, we computed the abundance of haddock in the 



Hayes, D., and N. Buxton. 1992 Assessment of the Georges 

 Bank haddock stock in 1991. Northeast Fisheries Science 

 Center. Res. Doc. SAW 1,3/1. (Appendix to CRD-92-02.) 



strata sections in the same manner as was done for 

 entire strata according to the method in Smith ( 1988). 

 Total abundance on either side of the ICJ line was 

 then obtained by summing the respective abundances 

 from strata or strata sections, or from both. Abun- 

 dance for each stratum or stratum section was ob- 

 tained as the average catch per tow for tows within 

 it multipied by the number of possible tow units 

 within it, thereby weighting the average catch per 

 stratum or stratum section by its area. 



This approach could be applied in most instances; 

 however, there were cases in the NMFS surveys 

 where no tows were made in a stratum section ( Tables 

 1 and 2). Missing observations were handled in three 

 ways; by estimation using the multiplicative model, 

 by assumption of a zero value, or by assigning the 

 mean per tow obtained for the whole stratum. The 

 multiplicative model (Gavaris, 1980) was the pre- 

 ferred method but it requires a In transformation of 

 the catch per tow from the surveys which precludes 

 the use of "0" values in the model; therefore, for strata 

 sections with low abundance, this model was inap- 

 propriate. The U.S. sections of strata 17 and 18 in 

 both spring and fall surveys, the U.S. section of stra- 

 tum 19 in the fall, and the Canadian section of stra- 

 tum 18 in the spring had a predominance of "0" val- 

 ues, suggesting that it would be reasonable to as- 

 sume a zero value for missing observations in those 

 strata sections. Missing observations for U.S. stra- 

 tum section 20 within 5Zj,m were assumed to be 

 equal to the mean catch per tow obtained for the 

 entire stratum that extends west of 5Zjm, and the 

 age composition for the whole stratum was extrapo- 

 lated to that portion lying within 5Zj,m. This proce- 

 dure was first followed for the 1991 eastern Georges 



