Laboratoire des Écoulements Géophysiques et Industriels




Nos tutelles

CNRS

Nos partenaires

Rechercher


Accueil > Actualités > Séminaires > Séminaires 2007

Jeudi 13 septembre 2007 14h

Francesco d’Ovidio (Laboratoire de Météorologie Dynamique - École Normale Supérieure - Paris - France)

Titre/Title :
Linking horizontal stirring to biogeochemical tracers

Contact :
Jacques Verron

Résumé/Abstract :

In the ocean as in the atmosphere, remote and in situ tracer observations show the presence of strong gradients occurring at the mesoscale and below, often dificult to explain with available instantenous velocity snapshots. This is particularly evident in the ocean, where tracers like surface temperature and chlorophyll are typically organized in filamental structures covering the 1-10 km scale, while velocity datasets (from altimetry or assimilation) show mesoscale structure in the range of 50-100 km. Due to their extremely high computational cost in space and time resolution, filaments cannot be explicitely represented in global circulation models and therefore their dynamics is currently a major uncertainty in the determination of global biogeochemical budgets. Fousing on the Lyapunov technique and real data, I will show how submesoscale tracer patterns can be reconstructed by using coarser velocity fields if both the spatial and time variability of the velocity is taken into account. Transport barriers with the typical lobular structure observed in tracer fields can be systematically derived. Comparison of transport barriers with tracer fronts can be used to quantify the effect of horizontal stirring on tracer varibilities as well as an objective validation criterion for comparing velocity fields of different products. Moreover, the calculation of Lyapunov exponents is also able to provide information on the mixing associated to the filamentation process, suggesting a way for parameterising filamentation events into an equivalent diffusion and providing a tool for studying its regional and global variability.