218 
Fishery Bulletin 96(2), 1998 
log fishing in response to pressures from the canner- 
ies, the MMPA, and the IDCA. Because all large ETP 
purse-seiners use the same type of gear, fishing pro- 
cedures, and vessels, a rough estimate of tuna dis- 
card for the entire international fleet can be obtained 
by simply applying the ratio of discard of total catch 
for the U.S. fleet to total catches reported for the in- 
ternational fleet, under the assumption that non-U. S. 
fleets discarded tuna in the same manner as the U.S. 
fleet 9 during the study period. 
Average annual commercial catch (short tons 
landed, all tuna species combined) by U.S. vessels 
fishing in the ETP during 1989-91 was 76,425 short 
tons. 10 Estimated average annual discard for the U.S. 
fleet during the study period was 6,080 short tons. 
Thus, estimated total annual discard by the U.S. fleet 
under observed patterns of effort was 7.96% (6,080/ 
76,425 x 100) of the average annual commercial catch 
(Hall * 11 ). As noted previously, predicted total annual 
discards from the U.S. fleet for redistributions of dol- 
phin sets to log or school sets (or both) increased pre- 
dicted total discard by 1.25 times for redistribution 
to school sets, 2.31 times for redistribution to a com- 
bination of log and school sets, and 3.37 times for 
redistribution to log sets (Fig. 5). 
Average annual commercial catch for the interna- 
tional fleet during the study period was 395,768 short 
tons. 12 If observed effort patterns of the U.S. fleet 
9 Although we cannot test this assumption at present, presum- 
ably all vessels selling fish to canneries serving international 
markets have the same constraints on acceptable sizes for in- 
dividual tuna retained for sale. It is possible that smaller sizes 
may be acceptable by canneries located in individual countries, 
but we are not aware of any such policy. The implications for 
population dynamics of affected tuna will be the same regard- 
less of whether the fish are marketed or not (they are removed 
from the population in either case). 
We also assume here that the effort distribution is the same 
for both the international and U.S. fleets. That is not strictly 
true, but the main difference tends to be that the international 
fleet, composed mostly of Mexican vessels, fishes more inshore 
areas (i.e. more in area 1) than does the U.S. fleet alone. Fish- 
ing more inshore would increase the fraction of school and log 
sets in the data. Thus, extrapolation from the U.S. data will, if 
anything, underestimate the extent of bycatch on school and logs. 
10 Estimated annual catches of tuna (all species combined, sur- 
face fishing only) for the U.S. fleet were 110,981, 79,530, and 
38,765 short tons for the years 1989, 1990, and 1991, respec- 
tively ( IATTC, 1991 ), averaging 76,425 short tons annually for 
the period. 
11 This compared well with IATTC data showing 7.4% of all 
sampled tuna captured during 1993 and 1994 were discarded 
(13,500 short tons of tuna discarded from 181,422 short tons 
of tuna caught) (Hall, M. 1995. Inter-American Tropical 
Tuna Commission, do Scripps Institute of Oceanography, La 
Jolla, CA 92037. Unpubl. data). 
12 Estimated annual catches of tuna (all species combined, sur- 
face fishing only) for the international fleet were 441015, 
407445, and 338,385 short tons for the years 1989, 1990, and 
1991 respectively (IATTC 5 ), averaging 395,768 short tons an- 
nually for the period. 
during the period were the same as effort patterns 
of the international fleet, then discard from the in- 
ternational fleet under hypothetical redistributions 
of dolphin effort can be simply estimated as 7.96% of 
395,768 short tons, i.e. 31,500 short tons. Assuming 
that changes in effort would affect discard in the same 
way for the international fleet as for the U.S. fleet, 
then discard under hypothetical redistributions of 
dolphin effort can be predicted as 1.25 times, 2.31 
times, and 3.37 times 31,500 short tons, for redistri- 
butions to all school sets, combined school and log 
sets, and all log sets, respectively. The predictions 
are then 39,200, 72,700, and 106,000 short tons of 
tuna discarded, respectively (Fig. 6). 
The significance of these discard predictions for 
future tuna resources depends on their magnitude 
in relation to the total stocks of tuna in the ETP. Al- 
though the amount of tuna that is potentially dis- 
carded appears high, it will not be particularly im- 
portant if it represents only a few percent of the to- 
(reported) Estimated Case 1 Case 2 Case 3 
Figure 6 
Average annual commercial tuna catch and estimated av- 
erage annual tuna discard by the U.S. and international 
purse-seine fleet fishing in the eastern tropical Pacific 
Ocean, 1989-92, under observed effort patterns and redis- 
tributed effort patterns where all dolphin sets were redis- 
tributed exclusively to school sets (case 1 ), equally to school 
and log sets (case 2), and exclusively to log sets (case 3). 
