Sladek Nowlis and Roberts: Fisheries benefits and optimal design of marine reserves 



61 



only sustain modest harvesting ef- 

 fort (Fig. 31. In contrast, life history 

 parameters from the literature sug- 

 gested that red hind had a relatively 

 high A = 1.31. Consequently, its 

 maximum sustainable yield oc- 

 curred at the highest fishing mor- 

 tality of any species we tested (Fig. 

 3). The two other species we exam- 

 ined had intermediate intrinsic 

 rates of population growth rates and 

 responses to reserves. 



The sensitivity of our models' 

 quantitative predictions was also 

 clear within a species when we var- 

 ied larval survivorship. For all spe- 

 cies, optimal reserve proportion and 

 yield without a reserve varied 

 greatly (senior author's unpubl. 

 data) because we varied larval sur- 

 vivorship from 10"^ to 10"*. This 

 sensitivity to poorly understood pa- 

 rameter values renders any quan- 

 titative estimates of optimal reserve 

 proportion unreliable, whether the 

 inaccuracy is in larval sui-vivorship, 

 the relationship, or parameters for 

 density dependence, or any other 

 life history parameter. 



Finally, we examined how re- 

 serves might influence unpredict- 

 able catches resulting from environ- 

 mental variation. Our stochastic 

 models predicted that catches will be more stable 

 with larger resei-ve proportions. In these models, we 

 saw general decreases in catch variability with in- 

 creasing reserve proportion (Fig. 5). The results pre- 

 sented here showed drops in variation that were more 

 pronounced at higher fishing mortalities for all four 

 species. We also tested these results at three levels 

 of environmental variation. Our results showed that 

 the drop in catch variability was most extreme when 

 the environment was most variable, suggesting that 

 the stability offered by reserves will be most valu- 

 able in highly variable fisheries. 



Discussion 



Effects of life history and 



fishing mortality on reserve benefits 



Our models predicted that marine fishery reserves 

 will provide catch enhancements to any overfished 

 fishery that meets our basic assumptions regarding 



Reserve proportion (s) 



Figure 5 



Catch variability and reserve size. (A) Panulirus penicillatus. Red Sea spiny 

 lobster. (S) Balistes vetula. the queen triggerfish. IC) Haemulon plumien, white 

 grunt. iD) Epinephelus guttatus, red hind. Each graph shows decreasing catch 

 variability with increasing reserve prnportion at four levels of fishing mortality. 



the movement of adults and larvae. The results from 

 previous modeling efforts by Man and colleagues 

 (1995) and Holland and co-workers (Holland and 

 Brazee, 1996; Holland et al.') support these findings 

 if one compares their results in specific cases to the 

 patterns we found for a variety of species. Two key 

 variables help determine whether a population is 

 overfished: intrinsic population growth rate (A) and 

 fishing mortality. Managers can control fishing mor- 

 tality to varying extents. Apparently, this control is 

 inadequate in many industrial fisheries (FAO, 1995) 

 and is probably even less effective in subsistence fish- 

 eries (Roberts and Polunin, 1993). Managers have 

 no control over population growth potential but can 

 take into account that species with low population 

 growth have a greater tendency to be overfished and 

 consequently show greater promise for fisheries en- 

 hancements from reserves. 



Even in a well-managed fishery, it may be helpful 

 to close large areas. This strategy could allow the 

 relaxation of some fishing restrictions in remaining 

 waters. Consequently, recreational and commercial 



