Projecting uncertainty onto marine megafauna trajectories

In this study, a method is proposed for estimating the uncertainty of a Lagrangian pathway calculated from an undersampled ocean surface velocity field. The primary motivation and application for this method is the differentiation between active and passive movements for sea turtles whose trajectories are observed with satellite telemetry. Synthetic trajectories are launched within a reconstructed surface velocity field and integrated forward in time to produce likely trajectories of an actual turtle or drifter. Uncertainties in both the initial conditions at launch and the velocity field along the trajectory are used to yield an envelope of possible synthetic trajectories for each actual trajectory. The juxtaposition of the actual trajectory with the resulting cloud of synthetic trajectories provides a means to distinguish between active and passive movements of the turtle. The uncertainty estimates provided by this model may lead to improvements in our understanding of where and when turtles are engaged in specific behaviors (i.e. migration vs. foraging)—for which potential management efforts may vary accordingly.

► We demonstrate the impact of uncertainty on analyses of sea turtle tracks.
► We model the uncertainty of Lagrangian pathways from undersampled velocity fields.
► We develop a Monte-Carlo model for creating “daisy-chain” trajectory clouds.
► We differentiate between active and passive sea turtle movements.