Open tdjames1 opened 2 years ago
The actual release times for each day are available in the files here:
/nfs/see-fs-02_users/lecsra/MOSAiC/DRIFT/trelease
The number of release points in these files should correspond with the number of trajectories for each day. Remember that the trajectory filenames have a date of midnight (00) the day following the arrival times on the ship. This is remedied in the interpolated output trajectories, where I have renamed the files to be those of the day of ship data (rather than midnight 00 the following day).
Some of the model output files do not contain the full number of trajectories (1440).
For these files, assigning a sequential date index to trajectory data is likely to be incorrect since presumably missing trajectories could occur at any point during the 24 hour period covered.
According to Steve, missing trajectories are due to "missing ship data". Is there a way to determine which trajectories are missing?
Exact time and location of the ship are available in a file for each day: /nfs/see-fs-02_users/lecsra/MOSAiC/DRIFT/trelease
Timestamp on the files is one day different from model output files
e.g. missing trajectories in output for 20200127:
correspond to missing data in release log for 20200126:
Hence, need to use the release/ship data to build a datetime index for the trajectory data.