FISHERY BULLETIN: VOL. 79, NO 2 



Rn-l - ^n 



F , A . 



n-l M - 1 



(9) 



By repeating this procedure, Equations (9) and (8), 

 using Nn -1 as an estimate of 2-yr-olds alive at 

 the beginning of a fishing season, the number of 

 tagged 1-yr-olds (Nn-2) alive at the beginning 

 of a fishing season was estimated. 



Age ( n ) specific exploitation rates w^ere esti- 

 mated for each release area by using the estimated 

 age specific mortality rates in; 



F A 



n n 



(10) 



tagged fish to other causes, which is included in 

 the estimate of M , may be lower than for fish 

 tagged in all other areas. 



Estimates of initial tag loss from shedding and 

 death were made by comparing the number offish 

 estimated to have been alive at the beginning of 

 the first season with the number that should have 

 been alive if only natural mortality (M = 1.0935) 

 had caused deaths during the approximately 8 mo 

 (0.67 yr) following tagging. The apparent tagging 

 loss estimates (L) were calculated by estimating 

 the fraction by which M ' would have to be reduced 

 to equal A^i prior to undergoing 8 mo of natural 

 mortality, i.e.: 



(M'-LM')e-°-^^^ 



N, 



Except for fish tagged in the Mobile Bay, Ala., 

 area (Alabama coastal waters are closed to purse 

 seining) the exploitation rates for both age-1 and 

 age-2 fish declined progressively as the distance 

 from the delta increased. The decline was much 

 greater for age-1 than for age-2 fish (Figure 5). The 

 low rates of entry offish from the extremes of the 

 range and the purse seine closure imposed for 

 coastal Alabama waters may enable a small buffer 

 stock to survive in the event of heavy exploitation. 



For fish tagged in Mobile Bay, the number esti- 

 mated to have been alive at the beginning of the 

 first fishing season exceeded the number of fish 

 tagged, an obvious impossibility (Table 8). In out- 

 side waters, Gulf menhaden are taken inciden- 

 tally by the industrial bottom trawl fishery 

 (Roithmayr and Waller 1963) and by the shrimp 

 fishery in inside and outside waters. The over- 

 estimate of Ni may have been the result of esti- 

 mated M being too high for this group offish. One 

 possibility for M being too high is that the loss of 



z °* 



I 0.5 



< 



?: 0.3 



"- 0.2 

 O 



S! 0.1 



" o' J 1 1 1 1 1 i I ' 



1 234 5 6789 



RELEASE AREA 



Figure 5. — The rate of exploitation by release area for age-1 

 (dots) and age-2 (circles) Gulf menhaden tagged as juveniles in 

 estuaries on the Gulf of Mexico. Arrows indicate degree and 

 direction of change between years. 



by solving for L and simplifying, 



L = 1- 



M'(e" 



■0.61M 



) 



(11) 



The resulting estimates, expressed as a percent- 

 age of the number offish tagged, ranged from 22.1 

 to 63.0f7f (Table 9). These estimates seemed realis- 

 tic in view of the loss rates reported by Kroger and 

 Dryfoos (1972) for Atlantic menhaden of similar 

 size tagged with small (juvenile) tags. These esti- 

 mates do not, however, have any bearing on the 

 20% tagging loss estimate used for the adult study, 

 as this study used larger fish and larger tags. 



Table 9. — Numbers of juvenile Gulf menhaden tagged in 

 autumn, numbers estimated by sequential analysis to have been 

 alive the following April, and effectual tagging loss rates, Gulf of 

 Mexico. 



'Weighted data for this area only. 

 ^Unrealistic results, see text. 



ACKNOWLEDGMENTS 



Thanks are due to the many past and present 

 employees of the Southeast Fisheries Center 

 Beaufort Laboratory, NMFS, NOAA, who were 

 involved in the many aspects of the menhaden 



334 



