732 



Abstract— Estimates of instantaneous 

 mortality rates (Z) and annual appar- 

 ent survival probabilities (.0) were 

 generated from catch-curve analyses 

 for oceanic-stage juvenile loggerheads 

 (Caretta caretta) in the waters of the 

 Azores. Two age distributions were 

 analyzed: the "total sample" of 1600 log- 

 gerheads primarily captured by sight- 

 ing and dipnetting from a variety of 

 vessels in the Azores between 1984 and 

 1995 and the "tuna sample" of 733 log- 

 gerheads (a subset of the total sample) 

 captured by sighting and dipnetting 

 from vessels in the commercial tuna 

 fleet in the Azores between 1990 and 

 1992. Because loggerhead sea turtles 

 begin to emigrate from oceanic to neritic 

 habitats at age 7, the best estimates of 

 instantaneous mortality rate (0.094) 

 and annual survival probability (0.911) 

 not confounded with permanent emi- 

 gration were generated for age classes 

 2 through 6. These estimates must be 

 interpreted with caution because of the 

 assumptions upon which catch-curve 

 analyses are based. However, these are 

 the first directly derived estimates of 

 mortality and survival probabilities for 

 oceanic-stage sea turtles. Estimation 

 of survival probabilities was identified 

 as "an immediate and critical require- 

 ment" in 2000 by the Turtle Expert 

 Working Group of the U.S. National 

 Marine Fisheries Service. 



Estimates of survival probabilities for 

 oceanic-stage loggerhead sea turtles 

 {Caretta caretta) in the North Atlantic 



Karen A. Bjorndal 

 Alan B. Bolten 



Archie Carr Center for Sea Turtle Research and Department of Zoology 



University of Florida 



P.O Box 118525 



Gainesville, Florida 32611 



E-mail address (for K, A. B|orndal): kab@zoology ufl edu 



Helen R. Martins 



Deparlamento de Oceanografia e Pescas 

 Universidade dos Azores 

 PT-9901-862 Horta 

 Acores, Portugal 



Manuscript approved for publication 

 21 April 2003 by Scientific Editor 



Manuscript received 26 June 2003 at 

 NMFS Scientific Publications Office. 



Fish. Bull. 101:732-736(2003). 



A major gap in our understanding of 

 sea turtle demography is the level of 

 mortality — both natural and human- 

 induced — experienced by wild popula- 

 tions. Lack of directly derived estimates 

 of mortality (or survival) probabilities 

 for the juvenile oceanic-stage in sea 

 turtle populations is a critical source of 

 uncertainty in development of popula- 

 tion models and evaluation of man- 

 agement plans. In current population 

 models, survival estimates for juvenile, 

 oceanic-stage sea turtles are fitted 

 parameters, not directly derived esti- 

 mates, of survival (Chaloupka, in press; 

 Heppell et al., in press). The Turtle 

 Expert Working Group ( 2000) identified 

 the estimation of survival probabilities 

 as "an immediate and critical require- 

 ment." Population models indicate that 

 survival probability of juvenile oceanic- 

 stage loggerhead sea turtles (Caretta 

 caretta) has a substantial effect on 

 overall population growth (Chaloupka, 

 in press; Heppell et al., in press). 



Catch-curve analyses have been used 

 for many years to estimate survival 

 probabilities for species harvested in 

 commercial fisheries and, less fre- 

 quently, for other species (Seber, 1982). 

 Estimates of survival probabilities 

 have been generated from catch-curve 

 analyses for subadult neritic-stage 

 populations of loggerhead sea turtles 

 (Frazer, 1987; Epperly et al., 2001) and 

 Kemp's ridley sea turtles (Lepidochelys 

 kempii) (Turtle Expert Working Group, 



2000) based on stranding data. Catch- 

 curve analyses confound mortality and 

 permanent emigration, and thus gen- 

 erate estimates of apparent survival 

 probability (CP), 



= S (1 - emigration ), 



where S = true survival probability; 

 and 



emigration = the probability of perma- 

 nent emigration. 



We estimated survival probabilities 

 (both <X> and S) for juvenile oceanic- 

 stage loggerhead sea turtles in the wa- 

 ters around the Azores, a developmental 

 habitat for the population of loggerhead 

 sea turtles that nest on beaches in the 

 southeastern United States (Bolten 

 et al., 1998). We applied catch-curve 

 analyses to two age distributions of 

 loggerhead sea turtles. 



Methods 



Two size distributions were compiled 

 for this study The first ("total sample") 

 comprised 1600 oceanic-stage logger- 

 head sea turtles that were captured 

 from 1984 through 1995 in the waters 

 of the Azores. Except for a few of the 

 smallest of these sea turtles found as 

 stranded carcasses, they were collected 

 in dipnets after being sighted at the 

 surface of the ocean from the decks of a 



