assumed to come from the Niantic River, as the major impact of MNPS operations. The latest application 

 of the model was for the MNPS Unit 3 Environmental Report - Operating License Stage (NUSCo 1983c). 

 Based on data from the literature as well as from NUEL studies at the time of the model run, a potential 

 5 to 6% decrease in winter flounder abundance was projected to occur after 35 yr of plant operations. 

 The population would recover to within 1% of the equilibrium level after an additional 65-yr period. 



It was emphasized in NUSCo ( 1983c) that the above model results were probably conservative. The 

 effects of entrainment were overestimated since vertical stratification of larvae and vertical variations in 

 current velocity were ignored, as was the potential for a fraction of the entrained larvae to survive, and 

 because of input of larvae from outside the Niantic River. The immigration of winter flounder from other 

 stocks into the area; density-dependent growth, fecundity, adult mortality; and, in some cases, density- 

 dependent larval mortality were not considered. Criticisms could be made of many of the model assump- 

 tions, particularly the ones made for estimating the Ricker (1954) stock-recruitment function parameters. 

 Nevertheless, the model has been independently evaluated and found to be an acceptable, conservative 

 method of assessing MNPS impact (Gore et al. 1977; Thomas et al. 1978). 



Some mathematical representation of the recruitment process is common to many fisheries population 

 models. Since egg and larval survival is more dependent upon environmental factors than adult survival, 

 a prominant feature of recruitment data for many species is the large amount of variability present (Jones 

 1982). In spite of this, deterministic population models use recruitment equations which have constant 

 parameters and assume either no adult age-structure or stable age-structure implying equilibrium. However, 

 models that take into account the effects of natural variability in the recruitment process have become 

 available recently. A stochastic population dynamics model, based on work by Lorda (1982) and Reed 

 et al. (1984), is currently being modified at NUEL to incorporate detailed early life history and temperature 

 effects on larval survival. This model explicitly uses the natural variability of some key population 

 parameters and data collected specifically for the Niantic River winter flounder stock. In recent years, 

 increased emphasis has been placed on understanding the dynamics of the early life history stages. The 

 factors governing the production and mortality of larvae should be known before assessing the impact of 

 MNPS operations. In particular, the quantification of larval mortality and its partition into density-dependent 

 and -independent components can be expected to be critical for the realistic estimation of entrainment 

 effects. Model results will include a probabilistic risk assessment analysis, which should provide better 



116 



