OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES 



Not only did seasonal Georges Bank assemblages 

 maintain their temporal integrity over the periods 

 1963-78 in the fall and 1968-78 in the spring, but they 

 also appear to have retained their spatial configura- 

 tion for the most part as well. The results of this 

 study indicate that although changes in species com- 

 position and relative abundance occurred in varying 

 degrees in all the assemblages, they remained con- 

 tinuous in time and spaca 



Although many of the species on Georges Bank 

 are found in several assemblages, it appears that 

 each of the five groups has enough large-scale varia- 

 tion in biomass, species composition, and relative 

 abundance to make each of the assemblages uniqua 

 Also at least one or two dominant Georges Bank 

 species occupy each assemblage, for example, the 

 bulk of the haddock stock occurs in the Northeast 

 Peak group. Thus, even though some assemblages 

 changed dramatically in terms of species richness 

 and relative abundance, the spatial integrity of each 

 complex was preserved over tim& 



The energy budget of Georges Bank can serve as 

 a plausible explanation for the particular species 

 distributions we found. Georges Bank is a dynamic 

 ecosystem driven by a complex and unique nutrient 

 advection system. Its shallow topography and geo- 

 graphic location, with constant mixing of the water 

 column and lack of stratification, does not lead to 

 the usual nutrient limitation of primary productivi- 

 ty (Sutcliffe et al. 1976; Cohen et al. 1982). Instead 

 of the usual spring and fall pulse in primary produc- 

 tion, the region is characterized by high productivi- 

 ty from April to November. Yearly primary produc- 

 tion levels are as high as 450 gC/m^ per yr in the 

 shallow (< 100 m) zone of Georges Bank (Cohen et 

 al. 1982). This shallow mixed zone encompasses the 

 same area as the Shallow and Intermediate assem- 

 blages delineated in our cluster analysis results. The 

 area is dominated by fish that feed on invertebrates. 

 Primary prey items for these species include euphau- 

 siids, copepods, mysids, amphipods, and other ben- 

 thic invertebrates. This part of the ecosystem is fairly 

 closely tied to primary production, and its compo- 

 nent species may compete for food resources dur- 

 ing their early life history (Pitt 1970; Bowman 1981; 

 Overholtz 1982). Predation, too, may be an impor- 

 tant biological mechanism for determing trends in 

 this assemblage (Overholtz 1982). The other assem- 

 blages that we have described in this analysis occur 

 along the fringes of Georges Bank at the shelf-slope 

 interface These shelf break groups contain the ma- 

 jor adult demersal fish stocks found in the area with 

 the exception of yellowtail flounder. These peripheral 

 assemblages are dominated by large predators that 



are generally piscivorous, with little, if any, dietary 

 overlap (Langton and Bowman 1981). 



The gradient analyses suggest that about 25% of 

 the total variation in species biomass distribution can 

 be explained with the variables used in the study. 

 This result was surprising at first, since we felt that 

 the variables we used would explain much more of 

 the variation than this. However, considering the fact 

 that other important biological factors, such as 

 predation, fishing, competition, and food preferences 

 were not included in the analysis, it is probably a 

 realistic percentage Perhaps an analysis that in- 

 cluded the whole east coast would account for much 

 more variation because a wider range of conditions 

 would exist. 



Other studies that have successfully explained 

 species distributions usually occur in habitats with 

 very strong physical or chemical gradients, such as 

 mountain forests or estuaries (Whitaker 1967; Mcln- 

 tire 1973). Either the actual gradients were not 

 strong enough to explain more than a small percent- 

 age of the species distribution or those other factors 

 were more important. 



The questions of resilience and stability of demer- 

 sal fish assemblages that were defined and inves- 

 tigated in this study have implications for the 

 management of Georges Bank. This study provides 

 a useful conceptual framework for managing many 

 of the demersal fish stocks in this area. Not only 

 were stable zones with specific resident fishes 

 delineated, but they were present over the long-term 

 record. Species components of fall assemblages are 

 indicators of general distributions that represent the 

 location of major fish stocks during the productive 

 portion of the year. Long-term responses observed 

 in the Georges Bank community indicate the pro- 

 pensity for adjustment or resilience (Rolling 1973) 

 that a particular assemblage might have Assem- 

 blages on the periphery of Georges Bank might be 

 less susceptible to changes in species composition 

 and relative abundance because their component 

 species are less trophically linked. The Shallow 

 assemblage, on the other hand, appears to be par- 

 ticularly vulnerable to fishing and perhaps inter- 

 specific interactions. This type of knowledge will be 

 helpful for understanding changes in fish abundance 

 and community structure and for effectively man- 

 aging fishery systems in the future 



ACKNOWLEDGMENTS 



We would like to dedicate this paper to the men 

 and women of the Northeast Fisheries Center, 

 Woods Hole, MA, for without their long-term plan- 



519 



