Van Eeckhaute et a\. Movements of Melanogrammus aeglefinus determined from a population model 



677 



>■'■•..*♦/•?♦ •I^ 



tf-' 



1963-71 



No./tow 

 .0.001 



• 25 



• 50 



, #100+ 



Figure 12 



Fall distribution uf age 1+ haddock on Georges Bank from NMFS bottom-trawl surveys. Data have been aggregated for the three 

 time periods shown. Note that 1969 data are not included as these data were unavailable. 



1965, biomass peaked at 427,000 t due to the recruit- 

 ment of the "very strong" 1962 and the "outstand- 

 ing" 1963 year classes that continued to dominate 

 the population during the late 1960s and early 1970s 

 (Clark et al., 1982). There was a large decline, by an 

 order of magnitude, in the biomass of haddock on 

 Georges Bank during the late 1960s and the early 

 1970s to a low of 14,000 t in 1973 (Clark et al., 1982 ). 

 The fall distribution pattern during this period of 

 exceptionally high but rapidly decreasing abundance 

 indicates that haddock within 5Zj,m were more 

 widely dispersed over the bank from 1963 to 1971 

 (Fig. 6) than that which was observed subsequently. 

 Moreover, the ratios of relative abundance did not 

 strongly favor the Canadian side of the ICJ line dur- 

 ing 1963-71 (Figs. 4 and 5). This difference in distri- 

 bution was also evident in the spring distribution 

 pattern for ages 4 to 8 (Fig. 6) and in the spring ra- 

 tios of relative abundance that were generally less 

 than 0.5 during 1968-71 (Fig. 8). After this period, 

 the ratios, though variable, were usually above 0.5. 

 Polacheck'^ noted that haddock, since the late 1960s, 

 were almost never caught in research survey tows 

 in the southern and central portions of Georges Bank, 

 areas where they had been caught regularly during 

 the mid-1960s. 



The post-1971 ratio of relative abundance trends 

 for haddock persist to the present. During the late 

 1970s and early 1980s, the 1975 and 1978 year 

 classes dominated the population and since then 

 there have been several moderately strong year 

 classes, i.e. the 1983, 1985, 1987, and 1992 (O'Brien 

 and Brown, 1996; Gavaris and Van Eeckhaute, 1997 ). 



2 Polacheck. T. 1995. CSIRO, Division of Fisheries, P.O. Box 

 1538, Hobart, Tasmania 7001, Australia. Personal commun. 



Despite accompanjring fluctuations in haddock abun- 

 dance, the relative abundance pattern has not re- 

 verted to that observed from 1963 to 1971. Pola- 

 check,- using percent zero tows and patchiness to 

 explore spatial distribution of Georges Bank haddock, 

 also observed that when abundance increased with 

 the 1975 and 1978 year classes, the patterns in spa- 

 tial distribution did not resemble those observed from 

 1963 to the late 1960s. The distribution pattern dur- 

 ing that time period was influenced strongly by the 

 exceptionally large 1963 year class and may not be 

 typical of the stock when it was in a more stable state 

 as during the 1935-60 period. Recognizing the two 

 spawning components and the likelihood that had- 

 dock may display different spatial affinities depend- 

 ing on their origin, we hypothesize that the 1963-71 

 distribution patterns may have been the result of an 

 unusually greater contribution of recruitment from 

 the southwestern spawning component. On the other 

 hand, since 1985, the southwestern spawning com- 

 ponent has probably been depleted to a greater ex- 

 tent and may not be contributing to Georges Bank 

 production to the degree that it did from the mid- 

 1970s to mid-1980s. The depletion of the southwest- 

 ern spawning component is well illustrated by the 

 fall distribution in 5Ze of ages 1+ haddock for three 

 time periods, 1963-71, 1972-84, and 1985-95 (Fig. 

 12). Note that the almost continuous distribution of 

 haddock seen along the northern edge of the bank 

 during the first two time periods is not apparent 

 during the later time period. 



Since 1985, the relative abundance across the ICJ 

 line and the net migration rates have likely not been 

 strongly affected by variations in the abundance of 

 the eastern Georges Bank spawning component. 

 However, these attributes may get distorted by dis- 



