2) The second topic we discussed was about eddies. Hao created two files that will be useful for that. They are both at 

http://www.o3d.org/abracco/LabSea/Arrigo_files

One file is ssh.mat, which has sea surface height. For that, you would have to plot SSH and see if there is an eddy at a given place of interest. The second file, ow.mat, contains the calculation of the Okubo-Weiss (OW) parameter (see paper attached for explanation of the parameter). In the ow.mat file, there is also a variable called sstd (for spatial standard deviation), which is the spatial standard deviation of the OW parameter for each day. So, for any given day, you will have the spatial distribution of OW and one value of the spatial standard deviation (sstd). With that, you can set a threshold for eddies: any region where the OW is smaller than -0.2*sstd will be an eddy.

The easiest way to do this, I think, will be for you to compute the threshold Th= -0.2*sstd (this will have to be done for each day, since sstd will be different for each day), and then plot that contour Th on top of the dye distribution for that day. The contour will give you the boundary of the eddies, and it will be easy for you to see if there is dye inside an eddy or not. Note that the calculation involves computing multiple derivatives, so it is quite likely that the field of OW will be noisy (it helps to squint when looking at it :)  ). I haven't looked at the OW field - if it is "too" noisy, then it may be safer to use the SSH field.

3) Hao also saved on the same folder the distribution of temperature, salinity and density, so that you can plot profiles as we discussed on the call yesterday. The files are quite big, so we saved only data from May-Oct 2003 and May-Oct 2008. I think the outputs are also only for your region of interest (around your boxes 1-6). The files are called profiles_may_oct_2003 and profiles_may_oct_2008. It includes year, month, day, temp03 (or temp08), salt, dens, lat and lon. Hopefully you will be able to find a profile that you like in 2003 or 2008. If not, let us know and we can try to prepare similar files for other years (although I should say that Hao will be away next week).