Porch et al.: A catch-free assessment model with application to Epinephelus ita/aro 



97 



1950 1960 1970 1980 1990 2000 2010 2020 



Year 



1950 1960 1970 



1980 1990 



Year 



2000 2010 2020 



Figure 6 



Base model predictions of (A) spawning biomass of goli- 

 ath grouper (£. itajara) in southern Florida in relation 

 to the equilibrium level associated with a spawning 

 potential ratio of 50%, s/Sjq,.,, and (B) the apical fishing 

 mortality rate on goliath grouper (£. itajara), F^ .^i. 

 The lines with small dashes represent approximate 

 80% confidence limits and the dashed horizontal lines 

 represent the levels associated with an SPR of 50% . 



The sensitivity run with the alternate selection curve 

 also produced more optimistic results (Fig. 8). Inasmuch 

 as the model now attributes most of the fishing mortal- 

 ity to age classes well beyond the age at first maturity 

 (see Fig. 41, the spawning stock biomass is estimated 

 to have been reduced to a lesser extent (to about 10% of 

 virgin levels by 1990 as compared to 5%). Thus, other 

 things being equal, recovery requires less time. The lev- 



el of F, 



increased with the alternate selection curve 



because fewer age classes are affected by fishing. 



Discussion 



All of the model formulations examined depicted the 

 same qualitative patterns: escalating fishing mortality 

 rates and rapidly declining spawning biomass, particu- 

 larly during the 1980s, followed by a sharp decrease 

 in fishing mortality and strong recovery in spawning 

 biomass after the 1990 harvest ban. These trends are 

 remarkably consistent with the anecdotal observations 

 shown in Table 1 and Figure 5 as well as with the expert 

 testimony given during the SEDAR stock assessment 



review. The estimated rapid increase in fishing mortality 

 during the 1980s appears to reflect a real increase in 

 effort that occurred due to elevated demand and selling 

 prices (Sadovy and Eklund, 1999), as well as the wide- 

 spread use of the LORAN-C navigational system (which 

 made it easier for fishermen to relocate productive off- 

 shore shipwrecks). Thus, it seems safe to conclude that 

 the population was overfished at the time the harvest 

 ban was imposed and is currently undergoing a sub- 

 stantial recovery. Less clear is the extent to which the 

 population has recovered since the harvest ban. 



Using the base model, we estimated that the harvest 

 ban has reduced fishing pressure by more than 50%, 

 but probably less than 90% (Fig. 9). Thus, there is a 

 strong chance that the current fishing mortality rate, 

 although greatly reduced as compared to the 1980s, 

 remains greater than i^jo'; 'i-^-' above 0.05/yr). This in 

 turn translates into less than a 40% chance that the 

 population will recover to levels above SgQ,- within the 

 next 15 years. Several fishermen have testified that the 

 harvest ban is probably less than 90% effective because 

 goliath grouper are still taken illegally in places and 

 because animals caught and released in deeper water 

 often do not survive^; therefore this result does not ap- 

 pear unrealistic. 



More optimistic results, implying a 70% to 80% 

 chance of recovery within 15 years, were obtained when 

 the DeMaria index was excluded or when selection was 

 oriented more towards older animals. There does not 

 appear to be a strong a priori case for excluding the 

 DeMaria index in favor of the REEF and ENP indices. 

 Although the coverage is rather limited, the trends of 

 the DeMaria index are consistent with those of the ENP 

 index (with a suitable time lag) and with anecdotal 

 accounts of the trends in other areas." The issue of 

 selection is more vexing. It can be argued that the age- 

 composition data from the ENP creel census adequately 

 reflects the composition of the juvenile catch inasmuch 



