61 4. Results Over the 6 nesting seasons covered by this study (2017-2018 to 2022-2023), 51 monitoring field trips were conducted, for a total effort of 98 days and over 300 hours of beach survey. This protocol led to the census of 2186 tracks, including 1105 nests based on the presence of body pits over the 28 prospected islets. 4.1 Nest detection probability The loggerhead turtle nest tracks persistence estimated on the ‘GLS’ rookery was based on the survey of n=286 nests over the 2019-20 to 2022-23 period. Empirical data are presented in Figure 17-A. The confidence interval of predictions for each season overlaps nearly fully (not shown) and it confirms that season effect is not present in this dataset (Wald t-test for Season effect, Fisher combined p-value=0.11). Autocorrelation structure is very high with Phi=0.47. It can be seen in Figure 17-A that very few false positive or negative detections are present in the dataset (a false positive or negative occurs when a non-detected track was seen again in the next field session). The fitted detection probability is shown in Figure 17-B including its confidence interval based on 104 bootstrapping estimates. The fitted parameters of the logit model are Intercept=1.82 and Age=-0.08 which can be converted into S and A of the detection probability da: = 1 (1 + ((− 1⁄(4. ))( − )) ⁄ (5) With S=-(1/(4*Age))=3.04 and A=4*S*Intercept=22.23. The half-time life of a body pit on these beaches is 22.23 days (95% confidence interval (CI) from 1 to 71 days, 50% CI from 5 to 29).
RkJQdWJsaXNoZXIy MjE1NDI=