to February water temperatures, with a nar- 

 row, high-peaked curve found during warmer 

 years (low recruitment) and a broad, flatter 

 curve during colder years (high recruitment). 



15. Laboratory studies showed that larval growth 

 rates were dependent upon water tempera- 

 ture. These studies suggested that the opti- 

 mum temperatures for growth were interme- 

 diate (6.9-7. 5°C), with decreased growth oc- 

 curring at lower (5.4°C) or higher (10.8°) 

 temperatures. 



1 6. Yearly growth rates were estimated using field 

 data since 1983 at station C in the Niantic 

 River and since 1976 for entrainment data. 

 Estimated growth rates were consistent with 

 laboratory estimates, again showing that 

 growth was dependent upon water tempera- 

 tures. 



17. Post-larval young-of-the-year winter flounder 

 have been sampled at two stations in the 

 Niantic River since 1983. Densities at station 

 LR in 1987 were liigher than in previous 

 years. Smaller differences in growth were 

 noted between stations LR and WA during 

 1986 and 1987 than during 1984 and 1985. 

 Differences among years may have been due 

 to density-dependent growth, especially at 

 LR. Survival rates were very similar among 

 years, regardless of densities of young. 



18. The winter flounder was the second-most 

 abundant fish impinged on the traveling 

 screens at MNPS since 1976. Relatively few 

 specimens were impinged at Unit 2 during 

 the past 3 years due to declining abundance, 

 varying plant operations, and possible reduc- 

 tions related to the construction and opera- 

 tion of Unit 3. The installation of fish return 

 sluiceways at Units 1 and 3 lessened the im- 

 pact of impingement on the winter flounder 

 because it has good ( > 85%) survival when 

 returned to the water. Routine impingement 

 monitoring at Unit 2 was discontinued in 

 December 1987 upon agreement between NU 

 and CT DEP. 



19. To predict the long-term effects of larval en- 

 trainment, an impact assessment model for 

 winter flounder is currently under develop- 

 ment, which includes hydrodynamics and 

 population dynamics submodels. The func- 

 tion of each submodel is the estimation of 

 the fraction of total larval production lost to 

 entrainment at the plant and the measurement 

 of any resulting population changes, respec- 

 tively. A newer, more accurate and detailed 

 hydrodynamics submodel is under develop- 

 ment at MIT. Larval behavior will be sim- 

 ulated to correspond more realistically to ob- 

 servations made in the field. A stochastic 

 age-structured population submodel will in- 

 corporate the three-parameter stock- 

 recruitment relationship, which includes a 

 measure of compensatory mortality and the 

 introduction of realistic environmental vari- 

 ability. 



20. Results from both larval analyses and the 

 three-parameter stock-recruitment relation- 

 ship showed that year-class strength was re- 

 lated to events in the early life history stages, 

 with colder winters associated with better re- 

 productive success. Greatest winter flounder 

 mortality took place during Stage 2 of devel- 

 opment, during which density-dependent 

 mortality probably occurred. 



References Cited 



Adams, E., and D. Cosier. 1987. Numerical 

 models of coastal circulation and dispersion 

 with application to power plant discharges. 

 MIT Energy Laboratory Report. Dept. of 

 Civil Engineering, Massachusetts Institute of 

 Technology, Cambridge, MA. 



Arai, M.N., and D.E. Hay. 1982. Predation by 

 medusae on Pacific herring {Clupea harengus 

 pallasi) larvae. Can. .1. Fish. Aquat. Sci. 

 39:1537-1540. 



Bailey, K.M., and R.S. Batty. 1984. Laboratory 

 study of predation by Aurelia aurelia on larvae 

 of cod, flounder, plaice and herring: develop- 



218 



