# Meeting with Ana Vaz (08/26/09)

- Ana has computed FSLE in Paulo’s model as well as particles trajectories. The output is every 6 days so the velocity field is interpolated between two snapshots. Although high values of backwar-in-time FSLE should represent submesoscale barrier to transport, some particle trajctories do cross these “barriers”. The problem may be due to 1) the differences in the routines to compute FSLE on one size and trajectories on the other side but, most likely, due to 2) the coarse resolution and the choice of d0 –the distance that particles need to be apart before recording the time.
- Thus, one idea would be to use a high-resolution eddy field and compute FSLE (and FTLE) for a coarser and coarser velocity field: How good FSLE (and FTLE) are for the coarse resultion grid?
- Another idea is the vertical structure of FLSE: How do FSLE change with depth?
- Another idea is the study of the relationship between FSLE (and FTLE) and the vertical velocity field. Paulo’s model (that lacks the general circulation) could be used because the vertical velocity field is readily available. However, Yianli’s model is a HYCOM type of model, that is known to present difficulty in computing offline the vertical velocity field (why?).
- Finally, another idea would be to compute (using observations and models), FSLE (and FTLE) using the full velocity field and only the geostrophic velocity.