CoAuthors

Francesco d'Ovidio (LOCEAN Laboratory, Sorbonne Université (UPMC, Univ Paris 06)-CNRS-IRD-MNHN, 4 Place Jussieu, F-75005 Paris, France); Alberto Baudena (Sorbonne Université & CNRS, Laboratoire d’Océanographie de Villefranche-sur-Mer (LOV), 181 chemin du Lazaret, Villefranche-sur-mer, 06300, France); Milad Fakhri (National Center for Marine Sciences, National Council for Scientific Research (CNRS-L), Batroun, Lebanon); Yannice Faugère (Space Oceanography Division, CLS, Toulouse, France); Rosemary Morrow (Laboratoire d’Etudes en Géophysique et Océanographie Spatiales, Toulouse, France); Laurent Mortier (LOCEAN Laboratory, Sorbonne Université (UPMC, Univ Paris 06)-CNRS-IRD-MNHN, 4 Place Jussieu, F-75005 Paris, France); Georges Baaklini (LOCEAN Laboratory, Sorbonne Université (UPMC, Univ Paris 06)-CNRS-IRD-MNHN, 4 Place Jussieu, F-75005 Paris, France)

An altimetry-based Lagrangian approach can be constructed by integrating numerically the surface currents and by creating in this way a set of trajectories of virtual surface drifters. The Lyapunov exponent is a Lagrangian quantity aiming at identifying transport barriers in the ocean known as Lagrangian fronts or Lagrangian Coherent Structures (LCSs). Lagrangian fronts detection techniques in the ocean have been commonly used in ecological and environmental studies thanks to their capability of mimicking the horizontal stirring process induced by the mesoscale currents, and therefore their ability to predict part of the filamentation which any advected tracer undergo. However, as the tracer is stretched along a Lagrangian front, the latter also moves horizontally in the orthogonal direction to the stretching effect. Although considered less significant than the stirring behavior, such a lateral movement of a front close to a tracer is highly important especially when it comes to contaminants dispersion. Predicting today’s fronts displacement and their new positioning is conditioned by the availability of the future current velocity fields provided by assimilation models. However, the position of the filaments predicted by Lagrangian methods inherits any spatial error already present in these models. Here we present a novel method, based on a yet unexploited information, the Lyapunov vector, which allows the computation of the speed at which well-defined Lagrangian fronts will move in the near-future and so their new placement without the need for future current velocity fields, hence usable with altimetry for near-real time applications. We focus in particular on an oil spill case: the Sanchi accident occurred in the East China Sea. Our results shows that a short-term prediction (3-day horizon) of the fronts displacement is possible without any information on the future velocity field. We then present a global-scale analysis in which we map the regions whose fronts have the highest drifting speeds.