General structure of code

The general workflow here is:

Access data:

• Download the data from the Carthe website: https://data.gulfresearchinitiative.org/ Data can be found by searching for GLAD and LASER drifters.
• Conver the dat files to mat files. Code to do this is present in code_to_process_data.

Key calculations:

• Calculate the ocean depth at each points of the trajectory and add to data file. Done in trajectory_depth.m. This is used in the next step to only keep data that is in deep ocean.
• Convert trajectories to binned pairs. (Use the vectorized version) trajectories2binnedpairs_vectorized.m. Generates files with names like structure_pairs*.mat.
  • In ths past this was done using the script called trajectories2binnedpairs.m, which used loops and relied on a function called calculate_seperation_timeseries.m. This is very slow.
• Calculate structure functions (2nd and 3rd order) from the binned pairs, and store the mean values in SF_*.mat. Dones in pairtimes2SF.m
• Do the helmholtz decomposition of SF2. At the moment done using a function called helmholtz_decompose in plot_SFs.m. SF2 decomposed not saved anywhere.
• Generate many SF3 samples using bootstrapping. Done in binnedpairs2bootstrappedSF3.m (the copy of this script was created to be able to run in parallel since this code takes a long time.)
• Convert from SF3 to KE flux. Done in SF3_2KEflux_ls.m.

Plotting functions:

• trajectories_movie.m, trajectories_plot.m, LASER_trajectories_movie.m: plot trajectories on a map.
• plot_SFs.m: plots of SF2, the data distribution, pdfs of du, etc.
• SF3_2KEfluxm: plots of SF3 and associated calculations.