VaLighan and Prager: Decline in abundance of Pngrtjs pagnis off the soutfieastern United States 



363 





2r 



Primary 



-ft— Primary (Miss) 

 -**- Alternate 



4000 



3000 



2000 



Primary 



-ft— PrifiBry (Miss) 

 -e— Alternate 



B 



Figure 4 



(A) Annual estimated instantaneous fishing mortality rates (mean 

 F on ages 4-8), and (B) recruitment to age 1 from calibrated VPA 

 applied to red porgy off southeastern United States based on differ- 

 ent catch matrices and treatment of zeroes (alternative treatment 

 of zeroes as missing value is indicated as "Miss" after Primary! in 

 MARMAP hook-and-line CPUE. 



200-300 t/yr), but well below landings taken earlier (e.g. 

 >800 t in 1981 an(i 1982). Estimates of fishing mortality 

 rates in the recent period are about 2.7 to 3.6 times those 

 that could produce MSY (Table 8). Because benchmark 

 estimates are based on selectivity patterns estimated for 

 1992-96, their values are expected to change with the 

 recent introduction of higher minimum size limits. How- 

 ever, the general picture is unlikely to change. 



Surplus-production model 



It was not possible to estimate the shape parameter of the 

 generalized production model from these data; therefore, 

 we present results only for the logistic model. Fits of that 

 model with various assumptions about B,//^ were statisti- 

 cally equivalent (log likelihoods from -2.29 to -2.2.5 ), indi- 

 cating that the data-model combination is not informative 

 about relative biomass level at the start of the time series. 

 We present results under a range of assumptions centered 

 on B^/K = 0.75 (Table 9). The estimate of MSY was moder- 

 ately sensitive to the assumption on B/A', varying about 

 18.5'^ around the central estimate (Table 9); it was also 

 higher than our estimates from age-structured methods 

 I Table 8). Nonetheless, the bias-corrected 80*^/ confidence 



interval on the central MSY estimate (Table 9 1 and the 

 corresponding 50% interval (not shown) included the age- 

 structured estimates of MSY; therefore there was some 

 agreement between methods. 



Estimates of stock and fishery status expressed in rela- 

 tive terms (Table 9) were similar to those from the age- 

 structured analysis (Table 8) and like those estimates de- 

 picted a heavily overexploited stock. Stock biomass at the 

 start of 1998 was estimated as about 25% of the biomass 

 that can support MSY (Bmhy'- ^ result that was remark- 

 ably insensitive to the assumption on B^/K. Fishing mor- 

 tality rate in 1997 was estimated as about twice the fish- 

 ing mortality rate associated with MSY (Fj^jsy'- a result 

 only slightly sensitive to the assumption on B^/K. Equilib- 

 rium yield available in 1998 was estimated as about 44% 

 of MSY; this reduction was due to the stock's reduction 

 well below B^^^y- 



Estimates of relative stock and fishery status over time 

 displayed a pattern of increasing fishing mortality rate 

 and decreasing abundance over time (Fig. 11). It appears 

 that the biomass moved below B^sy '" about 1982 and 

 that the fishing mortality rate has been above F^j^y since 

 about 1978. Estimated confidence intervals on F/F^gy 

 are remarkable in encompassing a wide range that since 



