LENARZ and ZWEIFEL: INTERACTION BETWEEN LONGLINE AND SURFACE FISHERIES 



1) For the first 5 yr only longliners fished and 

 only in rows 5 to 8. 



2) For the next 5 yr, this longline fishery was 

 augmented with surface gear in all cells adjacent 

 to the coast. 



3) Next, exploitation by the surface gear was 

 expanded to include all cells for 5 yr. 



4) Finally, for the last 5 yr, age specific surface 

 fishing mortality was reduced by 75% for fish <2.5 

 yr of age because much of the surface catch of 

 yellowfin tuna in offshore areas of the eastern 

 Pacific comes from schools associated with por- 

 poise. Typically, porpoise schools contain few yel- 

 lowfin tuna <2.5 yr of age (Calkins 1965). 



Steps 1,2, and 3 resem.ble the sequence of events in 

 the eastern Atlantic fishery for yellowfin tuna. 

 Yellowfin tuna first were exploited in a significant 

 fashion by longliners in a 10° band along the 

 equator, then a nearshore surface fishery became 

 significant, and in recent years some exploitation 

 by surface gear in offshore areas has occurred. To 

 our knowledge, step 4 has not occurred in the At- 

 lantic. Age-specific fishing mortality rates similar 

 to those by surface gears estimated by Lenarz et al. 

 ( 1974) for the Atlantic yellowfin tuna fishery were 

 used (Table 5). The Ricker yield equation was used 

 to calculate yield for each time-area stratum. 



Total yields per recruit were calculated and are 

 shown in Figure 17. Yields per recruit are quite 

 similar for both recruitment models except near 

 shore, where yield per recruit was considerably 

 higher for the inshore recruitment model than for 

 the uniform recruitment model. The difference in 

 yield per recruit betweeti the two models decreases 

 slightly as time increases. Yield per recruit closely 

 approached equilibrium yield within 3 yr after a 

 change was made in the fishery. Total equilibrium 

 yield per recruit with an inshore surface fishery 



Table 5. — Estimates of age-specific F on an annual basis used 

 as baseline for simulation. See text for modifications of mortality 

 rates during simulation. 



Uniform RecruitmenI 

 Inshore Recruitment 



04- 

 



200- 



FlGURE 17. — Yield per recruit of hypothetical yellowfin tuna 

 fishery: (a) total, (b) longliners in all areas, (c) surface gear in all 

 areas, (d) longliners in cells 71 and 85, (e) surface gear in cells 71 

 and 85, (f) longliners in cells 69, 84, and 97, and (g) surface gear 

 in cells 69, 84, and 97. 



and longline fishery was about 179c higher than 

 with a longline fishery alone, 54% higher with a 

 uniform surface fishery than with only a longline 

 and inshore surface fishery, and increased by 9% 

 when F for small fish was reduced by 75% . Under 

 the assumption that the catchability coefficient is 

 independent of area, the surface fishery increased 

 its equilibrium yield per recruit about fourfold by 

 increasing its effort about 12-fold when it ex- 

 panded into offshore waters. The same action de- 

 creased yield per recruit to the longliners by about 

 55%r. 



We next examined the potential yield per re- 

 cruit to longliners in rows 5, 6, 7, and 8 by starting 

 a longline fishery with the age-specific F vector 

 multiplied by the scalar 0.3 and then multiplying 

 by 1.3 each year afterward. Yield per recruit ap- 

 pears to approach an asymptote of about 6 kg for 

 inshore recruitment and 5 kg for uniform recruit- 

 ment (Figure 18). The reduction in catch per re- 

 cruit per effort by fishing is not significantly af- 

 fected by choice of recruitment model. Even 

 though catch per recruit per effort at high levels of 

 effort was only about 20% of that at the beginning 

 of exploitation, overfishing in a yield-per-recruit 

 sense did not occur. Average size of fish in the 

 catch was not significantly affected by the re- 

 cruitment model, and decreased from about 50 to 

 30 kg with increased fishing effort (Figure 18). 



A simulation for an inshore surface fishery indi- 

 cated an asymptotic production curve with a 



821 



