The basic inputs required by the model are: 

 survival estimates for each larval stage, eggs, and 

 age-class; fecundity rates for each age-class; an 

 initial age-distribution for the adult population; 

 estimated parameters of the recruitment model; 

 estimates of the mean and variance of the water 

 temperature during February in the Niantic River; 

 and the estimated larval mortality due to plant 

 entrainment. Except for the latter input (see dis- 

 cussion of the larval dispersion model), we have 

 data for most of the other model inputs. However, 

 better estimates of egg and larval mortality are 

 still needed. 



Long-term impact assessment 



The larval dispersal and entrainment model to- 

 gether with the stochastic population model will 



allow us to simulate the long-term effect of larval 

 entrairmient at the adult population level under 

 a variety of scenarios chosen to describe prevailing 

 levels of entrainment, multiple or single sources 

 of winter flounder larvae, and various levels of 

 fishing effort. The final output from this simu- 

 lation scheme will consist of time-series of ex- 

 pected population sizes suitable for estimating 

 long-term averages and standard errors for direct 

 comparisons, or for applying probabilistic risk 

 analysis (PR A). In the latter case, projected pop- 

 ulation changes are expressed in terms of the 

 probability that a postulated change (or no change 

 at all) will occur in a specified number of years. 

 Although both analytical methods are acceptable 

 for impact assessment purposes, PRA methodol- 

 ogy has been favored in recent environmental im- 

 pact and risk assessment studies by the U.S. En- 



AGEO 



Eggs 



Larvae 

 Juveniles 



\. 



Natural 

 Mori. 



Age 

 1 



7 



Age 

 2 



J 



( ADDED ] 

 y Mort. J 



SIMULATED 



IMPACT 



Egg Production 



Age 



/ 



Age 



Oldest 

 Age 



Annual Catch (Fishing mortality) 



Fig. 37. Box and arrow diagram of the life-cycle simulation scheme in the population model. Transfers 

 within the population fi.e., aging and reproduction) are indicated by thick arrows and occur once a year. 

 Losses due to mortality are indicated by thin arrows and occur both daily (dashed) and yearly (thin solid). 



Winter Flounder Studies 



211 



