FLIERL AND WROBLEWSKl: WARM CORK CILF STRfc:AM RINCS 



to the continental shelf should have considerable in- 

 fluence on the abundance and distribution of shelf- 

 water larval fish and consequently on their recruit- 

 ment to the fishery. Our simple mathematical model 

 suggests a major effect of the ring-induced cross- 

 shelf flows is to cause decreases in the larval density 

 ahead of the ring where there is onshore flow. In 

 addition, water being drawn off the shelf in an 

 entrainment feature behind the eddy can carry lar- 

 vae away, constituting a significant loss to the shelf 

 population. 



Research scientists of the National Marine 

 Fisheries Service are currently investigating the fre- 

 quency of entrainment events and their impact upon 

 the fisheries (A. Friedlander^). Here we shall briefly 

 examine the published literature on ring distribu- 

 tions and larval fish surveys. We have used the 

 schematic charts of sea surface temperature pro- 

 duced by NOAA and ONR (Office of Naval 

 Research). All of these records are necessarily in- 

 complete due to cloudiness in the imagery and the 

 difficulties in inferring water motions from the sur- 

 face features. We therefore will make a somewhat 

 subjective estimate of ring importance by consider- 

 ing the persistence of various eddies and their close- 

 ness to the shelf. 



Figure 10 is a composite of the observed 

 MARMAP distributions of cod Gadus morhua, lar- 

 vae in the shelf region near Georges Rank during the 

 late winter and spring of 1977 (Smith et al. 1979). 

 We have placed on this figure the approximate loca- 

 tions of the warm core rings shown on the Experi- 

 mental Ocean Frontal Analysis Chart produced by 

 the U.S. Naval Oceanographic Office for this period. 

 Figure 11 shows the corresponding MARMAP distri- 

 butions for haddock, Melanogrammus aeglefiniis, 

 larvae. 



Four warm core rings had trajectory paths near 

 enough to Georges Bank during the spring spawning 

 period in early 1977 to have had significant impact 

 upon the larval fish distributions according to the 

 predictions of our model. In front of each ring shown 

 on Figures 10 and 11, the concentration of larvae 

 over the shelf appears to be relatively low, much as 

 our model suggests would occur in areas of ring- 

 induced onshore flow. There even appear to be 

 indications of entrainment of larvae off the shelf 

 behind rings 77E (Figs. 10c, lie) and 77 A (Fig. lid). 

 Unfortimately the MARMAP station grid does not 

 extend beyond the 200 m contour, so that this cannot 



be confirmed from the historical data. Extending the 

 MARMAP grid further offshore would be of great 

 benefit in assessing losses of larvae both by rings and 

 by other offshore transports. It is also not possible to 

 differentiate between patchiness from concentrated 

 spawning upon Georges Bank or other biological 

 causes and ring-induced variations. Again this makes 

 the relationship between theory and data difficult to 

 assess. 



Warm core ring 77 A (Figs. 10, 1 1) is an example of 

 a nearly stationary eddy, remaining adjacent to 

 Georges Bank from 30 March to 25 May 1977. But 

 according to our theory, ring 77E is potentially even 

 more dangerous, since it slowly sweeps down the en- 

 tire Georges Bank edge between 18 May and 6 July, 

 moving at 3-5 cm/s. Drifting groups of larvae could 

 have caught up to the back side of ring 77E and been 

 drawn off the shelf. 



The survival of larvae in 1977 was low, and the 

 year class was weak for both cod and haddock (Ser- 

 chuk and Wood 1981; Clark et al. 1982). In other 

 years also, there appears to be a relationship be- 

 tween the frequency of ring interaction with 

 Georges Bank during the winter and spring spawn- 

 ing season and subsequent year-class strength for 

 both cod and haddock stocks (Table 2). Cod spawn on 

 Georges Bank from December into May and haddock 

 spawn from February into June (Smith et al. 1979). 

 Therefore we have examined the 6-mo period from 

 January to June each year from 1975 to 1979 for 

 ring activity using the charts of Bisagni (1976), 

 Mizenko and Chamberlin (1979a, b), Celone and 

 Chamberlin (1980), and Fitzgerald and Chamberlin 



Table 2. — Apparent relationship between years of less fre- 

 quent WCR (warnn core ring) interaction withi Georges Bank 

 during the winter and spring spawning season and strong 

 recruitment of cod and haddock stocks. 



'A. Friedlander, Northwest Fisheries Center Narragansett 

 Laboratory, National Marine Fisheries Service, NOAA, R.R. 7A, 

 Box .522A, Narragansett, RI 02882, pers. commun. January 1983. 



325 



