688 



Fishery Bulletin 100(4) 



the abundance of predators or competitors, thereby chang- 

 ing the ratio of predators to prey in the trophic chain. 



That the bycatch of Istiophoridae represents less than 

 0.021% of the total tuna catch and less than 10% of the to- 

 tal catches of billfishes currently reported for the eastern 

 Atlantic Ocean (assumed to fluctuate around 7000-8000 

 t per year) suggests that the direct impact of the purse- 

 seine fishery on these stocks is weak. By comparison, pre- 

 vious research has shown that the discards of small tunas 

 and the total bycatch (billfishes, sharks, other fishes, etc.) 

 generated by this fishery were close to 2% and 1.9%, re- 

 spectively (Ai-iz and Gaertner, 1999). Compared with the 

 longline fishery, the European tuna purse-seine fishery 

 generates less bycatch of billfishes than the longline fish- 

 eries targeting tuna (Matsumoto and Miyabe, 2000; Gon- 

 zalez Ania et al., 2001), swordfish (Mejuto et al., 2000), or 

 both species (Cramer, 2000; Marcano et al, 2000). 



One of the more general implications of our findings 

 concerns the impact of the ban of FADs by the purse-seine 

 fishery on the bycatch of Istiophoridae. Our analysis sug- 

 gests that this moratorium led to a decrease in inciden- 

 tal catch of marlin from 600-700 t to less than 300 t. In 

 contrast, this trend was reversed for sailfishes, but the 

 corresponding bycatch increased only from 25 t to 45 t. 

 Because in the present study we did not take into account 

 different probabilities (see Table 1 ) for each strata, it could 

 be argued that the Monte Carlo simulations lead to only a 

 partial exploration of the uncertainty in the calculation of 

 the total billfish bycatch. However, it would be interesting 

 to consider this source of uncertainty in the future. Conse- 

 quently, the potential for possible regulations at different 

 spatial and temporal scales needs further exploration. 



Large bycatches of billfishes could affect the food web of 

 the epipelagic ecosystem inhabited by other apex preda- 

 tors. However, the "zero bycatch solution" propounded by 

 some environmentalist groups could accelerate the change 

 in biomass ratios between the different trophic levels of 

 the ecosystem. In a critique of the conventional risk fac- 

 tors (e.g. biological reference points) used to define the risk 

 of extinction in marine fishes, Musick (1999) introduced 

 other interesting criteria, such as rarity of a species, the 

 small distribution range of a species, endemic species, and 

 specialized habitat requirements. As can be seen in Fig- 

 ures 1-4, the range of spatial distribution of the billfishes 

 is very large. It could be argued that billfishes are rela- 

 tively widespread but occupy very specific habitats within 

 their range, and as a consequence, habitat loss could be ex- 

 amined as a risk factor. However there is no clear evidence 

 to support this hypothesis for billfishes. 



The difficulty of objectively measuring an ecological risk 

 when it concerns unexploited components of the ecosystem 

 must be stressed because of the vagueness of this concept 

 (Antoine et al., 1998). As a consequence, estimating the 

 ecological risk is reduced to the analysis of the impact of 

 a fishing practice on a limited number of symbolic species 

 (e.g. dolphins, sea turtles; Hall, 1996). Nevertheless, there 

 is no reason to believe that these charismatic species play 

 a larger role in the food web of the epipelagic ecosystem 

 than other targeted or nontargeted species. As shown by 

 Kitchell et al, ( 1999), there was no clear conclusion on the 



ecological role of apex predators (including billfishes) in 

 foods webs of the central North Pacific. Furthermore, if 

 a decision is made to reduce the ecological impact of the 

 bycatch in a given fishery, management actions cannot be 

 focused only on providing full protection for a single spe- 

 cies (Hall, 1996). Although everybody understands what 

 ■'ecosystem overfishing" means, Murawski (2000) high- 

 lighted the lack of consensus for defining this concept and 

 suggested the need for objective metrics that gauge prop- 

 erties associated with the main features of the ecosystem 

 (e.g. production, diversity, and variability). 



In addition, decision makers need to evaluate manage- 

 ment options that are both scientifically credible and eco- 

 nomically practical regarding the use of the ecosystems. 

 Because billfishes are sold on the local African fish market 

 (Romany et al., 2000), the effects of the fishing on the 

 ecological processes, as well as on human activities, must 

 be evaluated. Although in the past the regulatory process 

 did not account for sharing of gains nor the social costs 

 associated with fishing practices (Antoine et al., 1998 ), evi- 

 dence suggests that the ecosystem approach for managing 

 marine resources should also include these socioeconomic 

 considerations in a multicriteria analysis (Chesson et al, 

 1999, Sherman and Duda, 1999). 



Conclusion 



This study examined the bycatch taken by the European tuna 

 purse-seine fishei-y in the eastern Atlantic Ocean. Results 

 obtained from this fleet have been extrapolated to the entire 

 purse-seine fleet operating in the same areas of this ocean. 

 The main conclusion of this paper is that the direct impact 

 of the purse-seine fishery on the billfish component of the 

 epipelagic ecosystem is weak. Results of this study provide 

 additional information on the effect of the moratorium on 

 FAD fishing. Although some caution is required at this stage, 

 due to limitations of the spatial and temporal sampling cov- 

 erage, it was found that the ban on FAD fishing operations 

 led to a substantial decrease in marlin bycatch. 



The present state of knowledge allows us to reach only 

 preliminary conclusions. However, it should be borne in 

 mind that inadequate data can lead to the formation of 

 misguided policies. It is clear that detailed information 

 on bycatch is needed to counter the arguments of those 

 who propose total bans on some fishing practices. Conse- 

 quently tuna commissions must continue to pay attention 

 to the collection of bycatch statistics and must encourage 

 fishermen to report incidental catches in their logbooks, 

 at least by large taxa (e.g. billfishes, sharks, etc.). In order 

 to use accurate information for management purposes, we 

 recommend that data from regularly conducted observer 

 programs be used as part of any future research. 



Acknowledgments 



This research was funded in part by the European Com- 

 mission (DG XIV) research project n° 96/028: "A study of 

 the causes of the increase in the catches of bigeye tuna 



