530 



Fishery Bulletin 100(3) 



o 



CO 



o 



o 



CNJ 



o 

 o 



Mean 



Q10% 



Median 



Q90% 



Q99% 



Mode 



Model Type 1 



0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 0.5 



Relative biomass (Bo/K) 



Figure 8 



Isopleths for mean, median, mode, Qxir.^ Qgo';- ^""^ Q99", ten-year stock- 

 rebuilding times simulated with six model types (Table 1) for Georges 

 Bank yellowtail flounder. Isopleths for one statistic and all six model 

 types are shown in each panel. Isopleths of mean and modal rebuilding 

 times in results for models with uncertainty in Fv,j,.j. (model types 2, 4, 

 and 6) may be distorted (flat) at relatively high fishing mortality levels 

 because fishing mortality exceeds the simulated true F(,,,.j- in some 

 simulations, so that the simulated stock may never rebuild. 



of r (Cadrin, 1999; Applegate et al.'). For the six stocks, 

 starting biomass in 1999 ranged from 25% to 93% of B^^-y- 

 and estimates of F^gy ranged from 0.5 to 0.8. Estimated 

 rebuilding times averaged 3.5 years (ranging from 1 to 7 

 yr) for 50% probability of attaining B^j,.;. (NOWG^). There- 

 fore, the age-based simulations indicated that isopleths 

 based on deterministic biomass dynamic models generally 

 performed well for overfished New England groundfish 

 stocks. Brodziak et al. (2001) analyzed stock-recruit data 

 and concluded that "(1) time horizons for rebuilding will 

 be uncertain, owing to recruitment variability, (2) some 

 productive stocks (haddock, yellowtail flounder) have seri- 

 al correlation in recruitment and this may either enhance 



or diminish chances for stock recovery." Thus, results 

 with surplus production models, age-structured models, 

 and stock-recruit analyses highlight the fundamental 

 similarities between a wide range of modeling approaches 

 (Sissenwine and Shepherd, 1987). 



The best choice of stochastic simulation model for devel- 

 oping and evaluating rebuilding plans will depend on the 

 situation. Sainsbury (1993) concluded that models incor- 

 porating simple assumptions about population dynamics 

 were more appropriate for evaluating performance of con- 

 trol rules than models with more complex assumptions. 

 PFMC used a simple model incorporating environmental 

 effects on recruitment with useful results. However, Bell 



