short tows and never reaches 100% because tur- 

 tles may be captured at any time during the tow 

 and will survive if captured in the latter stages. 

 Tows shorter than 1 hour and longer than 6 

 hours, however, are relatively uncommon in com- 

 mercial shrimping operations. 



In the southern North Atlantic, the CPUE for 

 all turtles was strongly dependent on depth (Fig. 

 1). In depths >10 fathoms, turtle captures were 

 rare, even though, based on aerial surveys (Fritts 

 et al. 1983), turtles are distributed well offshore 

 in waters considerably deeper than 10 fathoms. 

 The strong depth dependency of CPUE may re- 

 flect the fact that the continental shelf is rela- 

 tively narrow along the southeastern seaboard, 

 and the fact that most shrimping occurs in waters 

 <10 fathoms. In the Gulf of Mexico, CPUE ap- 

 peared to be relatively constant over all depths 

 (Fig. 1). 



These estimates are conservative because only 

 offshore (outside the barrier islands) effort and 

 turtle captures were considered. 



It should be emphasized that trawl related tur- 

 tle mortalities are not confined to U.S. waters, 

 but occur on a worldwide basis. The same turtle 

 populations impacted in U.S. waters are also im- 

 pacted in territorial waters of other countries. In 

 the case of the Kemp's ridley which is believed to 

 be equally distributed in United States and Mex- 

 ican waters, Mexican trawlers may account for 

 mortalities similar to those of U.S. trawlers. To 

 effectively protect sea turtles, international coop- 

 eration is essential. 



Acknowledgments 



We thank all individuals who participated in 

 the collection of data aboard commercial vessels 



UJ 



Q. 



o 



0.16 -■ 

 0.15 - 

 0.14 - 

 0.13 - 

 0.12 - 

 0.11 - 

 0.10 - 

 0.09 - 

 0.08 - 

 0.07- 

 0.06 - 

 0.05 - 

 0.04 - 

 0.03 - 

 0.02 - 

 0.01 - 

 0.0 - 



a SOUTHERN NORTH ATLANTIC 

 D GULF OF MEXICO 



n n ^ 



JZL 



6 7 8 9 10 

 DEPTH IN FATHOMS 



11 12 13 14 15 15 + 



Figure 1. — Catch per unit effort (turtles/net hour) as a function of depth for captures in the southern 



North Atlantic and the Gulf of Mexico. 



Conclusions 



From our analyses, it is evident that significant 

 numbers of sea turtles are captured by commer- 

 cial trawlers in both the Gulf of Mexico and the 

 southern North Atlantic, and that over 20% of 

 these turtles are drowned in the trawl. We esti- 

 mate that 9,874 loggerhead, 767 Kemp's ridley, 

 and 229 green turtles may be killed annually. 



and those persons who managed each of the 

 projects. In particular, we appreciate the contri- 

 butions of Frederick Berry, Andrew Kemmerer, 

 Walter Nelson, Wilber Seidel, John Watson, 

 Charles McVea, Charles Roithmayr, and Butch 

 Pellegrin. Rick Minkler and Mark McDuff pro- 

 vided computer programming support, Velda 

 Harris tj^jed the manuscript, and Arvind Shah 

 provided statisical expertise. 



816 



