Introduction
For marine turtles, the indirect influence of a dynamic beach environment hasimportance for the success or failure of each nest. At high density nestingbeaches, intraspecific nest destruction may be at least as important least as important indetermining nest success as are other factors (eg. erosion, flooding, predation)(Bustard and Tognetti, 1969) . In addition, the conditions of nest incubationdirectly influence the secondary sex ratio because marine turtles havetemperature dependent sex determination (Pieau et al., 1994). However, thecombination of direct and indirect processes on the outcome of the sex ratio ofhatchlings is seldom considered.
To address this topic, we used a mathematical model to predict howdensity-dependent nest destruction would affect the sex ratio of hatchlings whenthe mean temperature at a beach (and therefore the primary sex ratio) variesduring the season. To test the model required time series data on intraspecificnest destruction, pivotal temperatures, and temporal changes in the thermalprofile of the beach. We collected these data in French Guiana at the highestknown nesting density for Dermochelys coriacea in the Atlantic (Fretey andGirondot, 1989).
In this region, the nesting speciesnesting species of decreasing numerical importance are:Dermochelys coriacea, Chelonia mydas, Lepidochelys olivaceaand sometimes also Eretmochelys imbricata. The Ya:lima:po beach is locatedat the estuary of the Mana and Maroni rivers near the Surinam border. This is themain nesting beach and 90-95% of the nests of French Guiana are deposited on that8 km beach. The nesting season extends from early March in the cooler rainyseason to middle August in the dry season. The thermosensitive period ofdevelopment for sex determination (TSP) is in the middle-third part of theembryonic development (Desvages et al., 1993). Previous studies have shown thatnests deposited with TSP within the rainy season are mostly masculinized andthose deposited within warm season are feminized (Rimblot-Baly et al., 1986).
Material and Methods
At the peak of the nesting season, nesting activity is intense and nestingfemales often excavate previous nests with living embryos. To quantifyintraspecific yintraspecific nest destruction, we estimated the frequency of females excavatingpreviously laid nests containing living embryos. This parameter was estimatedduring the three main months of the 1994 nesting season using nest data groupedby 15 day intervals. The maximum proportion of nesting females excavatingprevious nests was 23%. Leatherback females nesting on the Ya:lima:po beach werecounted or estimated by interpolation during all the 1994 nesting season. 44% ofthe season's nests were deposited while the TSP was in a masculinizing period.This value was not used as the primary sex ratio but as an index to calculate theconsequences of this destruction by nesting females.
The analytical model is relatively simple and is described graphically in Fig. 1.The parameters in the model were: XY, the effective size of the nesting beach,ie. the size of an ideal beach in which females nest with uniform probabilityalong both axes; k, the daily probability that a nest will be destroyed byabiotic or biotic factors, but factors, but not including the destruction by nesting females;Dk is the daily change in this parameter and it can be positive or negative; andd is the probability that an excavated nests still contains living embryos.
We fitted this function iteratively to find optimal values for the parametersthat best explained the observed distribution of intraspecific nest destructionduring the nesting season. We used two iterative methods in sequence. First, weused a genetic algorithm to obtain a first approximation for the group ofparameters and then used a steepest descent gradient method to fit each parameterindividually and thereby obtain an optimal combination of fitted parameters.
Results
We found that the d parameter had so little influence on the distribution ofnesting females excavating previous nests that the parameter could not beaccurately estimated from the initial model. However, k, Dk and XY could beestimated (k=0.03225, Dk=0.16 10-4, XY=7 240 m2).
The k parameter was relativeter was relatively high: more that 3% of the nests were destroyed perday and the rate increased slightly during the nesting season. The effective sizeof the beach (= uniform nesting distribution along both axes) was one tenth ofthe actual size (70 000Êm2). This result was consistant with theobservation that nests are grouped near vegetation and the upper region of thebeach was far most affected than other regions. We examined the effects of theseparameters on those nests which successfully incubated full term. Because the dparameter could not be estimated, in all subsequent calculations we varied theparameter from 0 to 1.
In Fig. 2 are the distributions of intact nests at the end of the incubation ford = O and 1 using the set of parameters minimizing the diffrence between observedand calculated frequency of females excavating previous nest. As a firstapproximation, both distributions are very similar. Less than 10% of the nestscontained living embryos at the end of the incubation time; this value ; this value agreedwell with previous estimations made in 1979 (Fretey and Lescure, 1979).Moreover, the proportion of intact nests with TSP in a feminizing period wasenhanced compared to the distribution of the nest on the beach. This was becausethe early nests laid within the rain season were more subject to destruction thanthe later nests laid within the dry season. The effect of nesting density wasnext investigated since this factor varies annually. We compared the frequency ofintact and destroyed nests from 3 successive years of varying nesting densityconcordant with those observed in French Guiana (15 000, 30 000 and 60 000 nestsduring the season).
The data indicated density-dependence in nest survivorship, i.e. an increasingfemale density resulted in increased intraspecific nest destruction. Even so, theavailable space in which to nest was far from saturated because the number ofintact nests continued to increase when the number of nesting females increased(not shown). Surprisingly, this phengly, this phenomenon did not produce a significantdensity-dependent regulation of the population size and was thus contrary to theoutcome proposed by Bustard and Tognetti (1969). However, we did find that a biastoward feminization of primary sex ratio was density-dependent. When the numberof nesting females was enhanced, the bias was stronger.
In summary, the proportion of intact nests at the end of the incubation wasaround 10% and the value was density-dependent at this high density beach.Nest-destruction by nesting female leatherbacks at this high density nestingbeach was of minor importance compared to destruction by other abiotic factors (kand Dk). However intraspecific nest destruction did produce a significant bias inprimary sex ratio relative to the temporal distribution of nesting femalesarriving through the season.
Bibliography
Bustard, H.R. and Tognetti, K.P. 1969. Green sea turtles: a discrete simulationof density-dependent population regulation. Science 163: 939-941.
Desvages, G..
Desvages, G., Girondot, M. and Pieau, C. 1993. Sensitive stages for the effectsof temperature on gonadal aromatase activity in embryos of the marine turtleDermochelys coriacea. Gen. Comp. Endocrinol. 92: 54-61.
Fretey, J. and Girondot, M. 1989. L'activité de ponte de la tortue luth,Dermochelys coriacea (Vandelli, 1761), pendant la saison 1988 en Guyanefrançaise. Rev. Ecol. (Terre Vie) 44: 261-274.
Fretey, J. and Lescure, J. 1979. Rapport sur l'étude de la protection des tortuesmarines en Guyane française. Notes sur le projet de réserve naturelle deBasse-Mana. Minist. Environ., Mineogr. 59 pp.
Pieau, C., Girondot, M., Desvages, G., Dorizzi, M., Richard-Mercier, N. andZaborski, P. 1994. Environmental control of gonadal differentiation. In Short,R.V. and Balaban, E. (eds.), The Differences Between the Sexes. Cambridge Univ.Press, Cambridge, U.K., pp. 433-448.
Rimblot-Baly, F., Lescure, J., Fretey, J. and Pieau, C. 1986. Sensibilité à latempgrave; latempérature de la différenciation sexuelle chez la tortue Luth, Dermochelyscoriacea (Vandelli, 1761); application des données de l'incubation artificielle à l'étude de la sex-ratio dans la nature. Ann. Sc. Natur. 13: 277-290.
Fig. 1. Organigram of the simulation. See the text for the definition ofparameters.
Fig. 2. Distribution of the non-detroyed nests ((light grey) for d=0 and (darkgrey) for d=1) in comparison to the distribution of nesting females during thenesting season (-) for k=0.03225, Dk=0.16 10-4, XY=7 240 m2. The frequency ofnon-destroyed nests with TSP within dry season (feminizing season) is 59% and 69%respectively for d=0 and d=1.r d=0 and d=1.