Improved female abundance and reproductive parameter estimates through subpopulation-scale genetic capture-recapture of loggerhead turtles

Abstract

Nest counts are often used as indices for nesting female abundance in marine turtle monitoring, but accurately interpreting nest count trends requires context on the scale of demographic connectivity and estimates of reproductive parameters. Weak nest site fidelity (NSF) relative to the scale of tagging effort may bias parameter estimates. The reproductive ecology of Northern Recovery Unit (NRU) loggerhead turtles (Caretta caretta) was assessed through subpopulation-scale genetic capture-recapture via clutch sampling from approximately 1000 km of coastline from Georgia to Maryland, USA (30.75–38.06°N and 75.24–81.45°W). Of 20,682 clutches recorded from 2010 to 2012, 20,222 sampled clutches were assigned to 5684 unique females through microsatellite genotyping. Approximately 73% of females detected laying multiple clutches deposited them within a 20-km span, suggesting the possibility of demographic structuring across NRU rookeries that warrants further investigation. Estimated clutch frequencies (ECF) generated from open robust design modeling were 4.28 (4.02–4.54) in 2010, 4.63 (4.45–4.80) in 2011, and 4.57 (4.28–4.77) in 2012, and were significantly higher than observed clutch frequencies. ECF generated from single-island data were biased low by 23–50% relative to those from regional genetic tagging. Among females that nested at least once on a physical tagging beach, 54% also nested elsewhere, with 81% of these “permanently” emigrating during the nesting season. This pattern of relatively strong NSF, but distributed across multiple nearby islands, confounded modeling of detection in single-island datasets and highlights the need for regional coverage for generating robust estimates of demographic parameters for marine turtles.