Genetic determination of tag loss dynamics in nesting loggerhead turtles: a new chapter in “the tag loss problem”

Abstract

Capture–mark–recapture studies that fail to account for the frequency and dynamics of marker loss risk generating biased demographic estimates. In this study, we used permanent multilocus genotypes (i.e., “genetic tags”) and a new enhanced tag loss model to quantify the tag loss dynamics for both passive integrated transponder (PIT) and Inconel metal tags applied to loggerhead turtles (Caretta caretta) nesting on Wassaw Island, GA USA. Our results indicate that tag loss is most likely to occur within the nesting season in which tags were applied and is maximal just after tagging (maximum likelihood estimates): 0.00098 PIT tags day−1 and 0.007 Inconel tags day−1. After that, PIT tag loss was negligible and Inconel tag loss remained low but constant at 0.00028 tags day−1, such that after 5 years, the probability of losing one PIT tag was 0.06 and losing at least one Inconel tag was 0.46. The use of genetic tags in this study makes these the first truly accurate estimates of PIT and Inconel tag loss for marine turtles, and the new model of tag loss described herein represents an important advancement in the analytical methods used to estimate and compare tag loss dynamics.