One beach amongst many: how weak fidelity to a focal nesting site can bias demographic rates in marine turtles

In migratory marine species, demographic estimates are often generated from capture-mark-recapture (CMR) studies conducted at terrestrial breeding sites. However, when logistical difficulties limit the geographic area of these surveys, demographic estimates are vulnerable to biases. We compared demographic rates generated from CMR data of nesting loggerhead turtles ( Caretta caretta ) collected between 2010 and 2017 at one focal site (Wassaw Island, Georgia, USA; 31.89° N, 80.97° W) with estimates generated from the same group of turtles but including all other nesting events from adjacent sites in Georgia, South Carolina, and North Carolina. We found that estimates of annual recruitment at the focal site were overestimated: each year, 29–45% putative first-time nesters at the focal beach had, in fact, nested on a different beach in a previous season. Estimates of clutch frequency and breeding frequency generated at the focal site were biased low and skewed towards values of one, while estimates for remigration interval were not significantly over- or underestimated. Additionally, estimates of annual and total population productivity in terms of clutches, eggs, and hatchlings were underestimated by more than half at the focal site. Our results show how weak fidelity to a focal nesting/tagging site can affect demographic estimates in marine turtle populations and highlight the need to reconsider estimates from other populations that might be vulnerable to similar biases.