NOTE Caillouet et al.: Sea turtle strandmgs and shrimp fishing effort in northwestern Gulf of Mexico 



713 



northwestern Gulf are the loggerhead Caretta 

 caretta and Kemp's Ridley Lepidochelys kempi 

 (Rabalais and Rabalais 1980, Thompson 1988, 

 Amos 1989, Whistler 1989). 



This study deals with monthly sea turtle 

 strandings along shorelines and shrimp fishing 

 effort seaward of shorelines in the northwest- 

 ern Gulf. Strandings are observed for the most 

 part on barrier beaches, so they can be sum- 

 marized in linear distance units of shoreline. 

 Shrimp fishing effort is reported as days fished 

 within spatial units represented by shrimp 

 statistical subareas and 5-fathom (fm, 9.1m) 

 depth intervals (Kutkuhn 1962, Patella 1975). 

 To test the null hypothesis, we paired monthly 

 strandings along segments of shoreline with 

 monthly shrimping effort within 5-fm depth in- 

 tervals in the adjacent offshore waters. This 

 was done because it was expected that the far- 

 ther offshore the shrimping took place, the less 

 likely sea turtles impacted by such shrimping 

 would reach the shoreline, due to combined ef- 

 fects of surface currents, winds, waves, tides, 

 action by scavengers (e.g., sharks) and decom- 

 position of turtle carcasses (Heinly et al. 1988, 

 Murphy and Hopkins-Murphy 1989, Shoop and 

 Ruckdeschel 1989, Whistler 1989). Also, it is 

 possible that temporal-spatial distributions of 

 sea turtles and shrimp fishing activities overlap 

 only within certain depth intervals (Magnuson 

 et al. 1990). 



Materials and methods 



Since 1980, sea turtle strandings along the coasts of 

 the southeastern United States have been compiled 

 by the Sea Turtle Stranding and Salvage Network 

 (STSSN, Schroeder 1989). Shrimp fishing effort 

 statistics in the Gulf have been compiled since 1956 

 (Kutkuhn 1962). Our analyses were based on data from 

 1986-89, including sea turtle strandings available from 

 the STSSN database at the National Marine Fisheries 

 Service's (NMFS) laboratory in Miami, Florida, and 

 shrimp-fishing effort data available from the NMFS 

 laboratory in Galveston, Texas. 



Schroeder (1989) described the STSSN and pro- 

 cedures used to document sea turtle strandings. State 

 coordinators review and verify the stranding data sub- 

 mitted by network participants, then forward them to 

 the NMFS laboratory in Miami, Florida, where the 

 database is maintained. The database is not indepen- 

 dent of the distribution of human-induced mortality fac- 

 tors, temporal-spatial coverage is rarely uniform, and 

 most beaches are surveyed by volunteers (Magnuson 



940 \ Louisiana 



Texas 



Gulf of Mexico 



Mexico 



Figure 1 



Boundaries of upper coast (shrimp statistical subareas 17-18) and lower 

 coast (subareas 19-21) and 5fm (9.1m) depth intervals of the north- 

 western Gulf of Mexico (see Kutkuhn 1962, Patella 1975). 



et al. 1990). To improve temporal-spatial coverage and 

 supplement voluntary coverage, the NMFS Galveston 

 Laboratory initiated year-round surveys along the 

 coasts of southwestern Louisiana and Texas in 1986 

 (Heinly et al. 1988). Four-wheel-drive trucks, off-road 

 motor cycles, or all-terrain vehicles were used to survey 

 accessible gulfside shorelines at least once per month, 

 from the Mermentau River, Louisiana, to the Texas- 

 Mexico border (Fig. 1). The National Park Service 

 surveyed the Padre Island National Seashore near Port 

 Aransas, Texas, and the U.S. Fish and Wildlife Ser- 

 vice and Texas Parks and Wildlife Department assisted 

 in surveying Matagorda I. near Port O'Connor, Texas. 

 Reconnaissance flights conducted at least once monthly 

 were used to search for stranded turtles on San Jose I. 

 near Corpus Christi, Texas. 



For our analyses, sea turtle strandings (all species 

 combined) in the northwestern Gulf of Mexico during 

 1986-89 were extracted from the STSSN database. 

 Records of turtles caught by various commercial and 

 recreational fishing methods were deleted. Also deleted 

 were strandings of head-started (captive-reared) sea 

 turtles, because their distribution is influenced to some 

 extent by where they are released (Manzella et al. 1988, 

 Fontaine et al. 1989). 



