Individual And Age-Dependent Variations In Clutch Frequency Among Hawksbill Turtles At Cousin Island, Seychelles: 1973-1992

Jeanne A. Mortimer1 and Roby Bresson2
1Caribbean Conservation Corporation, Gainesville, FL 32602; and Department of Zoology, University of Florida, Gainesville, FL 32611 USA
2Cousin Island, via Grand Anse, Praslin, Republic of Seychelles

The turtles of Cousin Island are probably the most intensely studied population of nesting hawksbills in the world. All nesting emergences have been monitored since 1971 and a tagging program has been underway since 1973 (Diamond, 1976; Garnett, 1978; de L. Brooke and Garnett, 1983; Frazier, 1984%; Mortimer, 1984; Wood, 1986; Mortimer and Bresson, 1994). In the present study we examine 19 years of data (1 973 t o 1992) and report on the relationship between clutch frequency and relative age. Many sea turtle biologists have assumed that fecundity increases with the age of the turtle. This belief is based on the observation that “remigrant” turtles (i.e. those that arrive at the nesting beach bearing tags or tag scars from previous seasons) tend to lay significantly more egg clutches during the nesting season than do the neophyte or “recruit” turtles (i.e. previously untagged turtles). We examined two measurements of clutch frequency: the numbers of nestings recorded during consecutive remigrations to the nesting beach; and the time intervals that separated successive nesting seasons.

A large proportion of the turtles tagged at Cousin Island between 1973 and 1992 were recorded during only a single nesting season (79%). Looking in retrospect over those 19 years we found that a turtle that arrived at the nesting beach bearing either a tag or a tag scar from a previous season was more than twice as likely to be intercepted in a subsequent nesting season as was a turtle that arrived untagged.

For those turtles that were recorded nesting during more than one season, we performed an analysis of covariance using “annual clutch frequency” as the dependent variable and “season” and “tagging efficiency” as the independent variables. The analysis demonstrated that there is no steady increase in clutch frequency over time. It is possible, however, that clutch frequency increases during the first several nesting seasons of a turtle and then later declines. (At the time of this writing, we are analyzing all the data gathered through the 1993-94 season to determine whether such a pattern exists.)

Among 104 Cousin hawksbills, we recorded 186 remigration intervals (i.e. the number of years separating successive nesting seasons of a turtle.) The two- and three-year intervals together account for 85% of all intervals recorded. To test for bias caused by tag loss we compared the remigration intervals of the total population with those of a subset of turtles that were identifiable by a single tag throughout a period of eight or more years. A Kolmogorov Smirnov test showed no significant difference between the two patterns of distribution, thus demonstrating that the predominance of the (relatively short) two- and three-year remigration intervals is not an artifact of tag loss.

At Cousin, individual animals have been monitored over periods of up to 16 years. When we examined the lengths of consecutive remigration intervals for individual turtles we found no statistically significant changes in interval lengths over time.

In light of the predominance of the two- and three-year remigration intervals, we find it surprising that such a large proportion of the turtles were encountered during only a single nesting season. Moreover, the numbers of clutches recorded during the first (intercepted) nesting season of the single season nesters was significantly lower than the numbers recorded for the multiple season nesters.

How can we explain such a large number of single season nesters? Some possible explanations include high tag loss and mortality from poaching. We know that both factors have been important at Cousin Island, especially during the 1970’s. During recent years the percentage of single season nesters has declined significantly (Mortimer and Bression, 1994). It is also possible that a combination of insufficient beach patrols at Cousin and imperfect site fidelity in the Cousin tu~rtles causes some nesting emergences to go unrecorded. However, data available on nesting site fidelity in Seychelles hawksbills suggest that this may not be an important factor.

Other possible explanations for the large numbers of single season nesters recorded at Cousin and elsewhere in the world are the following. There may be a difference in reproductive fitness among individual turtles. This possibility has already been discussed by Richardson (1 982) and Tucker and Frazer (1991 1. Another explanation may be that there is a high rate of mortality during the first nesting season. (Note that the females of some species of penguins experience high rates of mortality during their first nesting season. Those birds surviving the first season, howevelr, show good survival in subsequent seasons.) It is possible that turtles generally exhibit low rates of fecundity during their first nesting season. If the first nesting season is then followed by a particularly long interval before a second nesting season, high rates of tag loss could occur. And in fact, a long interval between the first and second nesting seasons would not be surprising considering the very long interval we know occurs between hatching and first nesting.

Some combination of the above hypotheses probably explains the phenomenon of the “single season nesters.” But, this mystery will probably not be solved entirely by studies on the nesting beach. More in-water work is needed. More subadult turtles need to be followed into adulthood and more adult animals need to be tracked after they leave the nesting beach.

ACKNOWLEDGMENTS Funding for this study was provided by the International Council for Bird Preservation (ICBP), by the World Wide Fund for Nature (WWF-International) (Projects No. 1809 and MY0034lSC0009) and the Smithsonian Institution. We are also grateful for assistance from the U.S. National Marine Fisheries Service, Caribbean Conservation Corporation and the Archie Carr Center for Sea Turtle Research.

LITERATURE CITED de L. Brooke, M. and M.C. Garnett. 1983. Survival and reproductive performance of hawksbill turtles Eretmochelys imbricata L. on Cousin Island, Seychelles. Biol. Conserv. 25: 161 -1 70.

Diamond, A.W. 1976. Breeding biology and conservation of hawksbill turtles, Eretmochelys imbricata L. on Cousin Island, Seychelles. Biol. Conserv. 9: 199-21 5.

Frazier, J. 1984. Marine turtles in the Seychelles and adjacent territories. p. 41 7-468. In: Stoddart, D.R. (ed.), Biogeography and ecology of thle Seychelles Islands. Dr. W. Junk Publishers, The Hague, Netherlands.

Garnett, M.C. 1978. The breeding biology of hawksbill turtles (Eretmochelys imbricata) on Cousin Island, Seychelles. London, International Council for Bird Preservation (unpublished report). 33 pp.

Mortimer, J.A. 1984. Marine turtles in the Republic of Seychelles: status and management. ISBN 2-88032-901-9. International Union for Conservation of Nature and Natural Resources Publication Services, Gland, Switzerland, vii + 80 pp., 4 pis.

Mortimer, J.A. and R. Bresson. 1994. The hawksbill nesting population at Clousin Island, Republic of Seychelles: 1971 -72 to 1991 -92. p. 1 15-1 11 7. In: Schroeder, B.A. and B.E. Witheringon (compilers), Proceedings of the Thirteenth Annual Symposium on Sea Turtle Biology and Conservation. NOAA Tech. Mem. NMFS-SEFSC-341.

Richardson, J.I. 1982. A population model for adult female loggerhead sea turtles (Caretta caretta) nesting in Georgia. Ph.D. Dissertation, University of Georgia, Athens, Georglia.

Tucker, A.D. and N.B. Frazer. 1991. Reproductive variation in leatherback turtles, Dermochelys coriacea, at Culebra National Wildlife Refuge, Puerto Rico. Herpetologica 47(1 ):I 15-1 24.

Wood, V.E. 1986. Breeding success of hawksbill turtles Eretmochelys imbricata at Cousin Island, Seychelles and the implications for their conservation. Biol. Conserv. 37:321-332.

Paper presented at 14th International Symposium, 1994