NOAA-EMC / EVS

12 stars 26 forks source link

bugfix/subseasonal_avg_check #549

Closed malloryprow closed 2 months ago

malloryprow commented 2 months ago

Note to developers: You must use this PR template!

Description of Changes

The subseasonal CFS and GEFS grid2grid stats jobs failed due to the code trying to create a weekly averafe OSI-SAF file and was trying to use a file that didn't exist. This fixes the file checking and what files are used in the averaging for weekly and monthly OSI-SAF and GHRSST-OSPO files.

Related to #548

Developer Questions and Checklist

Testing Instructions

Set-up

  1. Clone my fork and checkout branch bugfix/subseasonal_avg_checks
  2. ln -sf /lfs/h2/emc/vpppg/noscrub/emc.vpppg/verification/EVS_fix fix

:black_square_button: subseasonal stats

  1. In versions/run.ver, uncomment export evs_ver=v1.0.12.
  2. cd dev/drivers/scripts/plots/subseasonal
  3. module reset
  4. Run the following: jevs_subseasonal_cfs_grid2grid_stats.sh, jevs_subseasonal_gefs_grid2grid_stats.sh
    • Set HOMEevs to the location of the clone.
    • Set COMIN to /lfs/h1/ops/prod/com/evs/v1.0/prep/subseasonal/atmos
ShannonShields-NOAA commented 2 months ago

@malloryprow and @PerryShafran-NOAA I have reviewed the util.py script changes and confirmed that the relevant sections include fixes (weekly and monthly osi as well as weekly and monthly ghrsst).

PerryShafran-NOAA commented 2 months ago

@malloryprow @ShannonShields-NOAA @AliciaBentley-NOAA The jobs have been submitted. The .o files will eventually be found here: /lfs/h2/emc/vpppg/noscrub/perry.shafran/pr549test/EVS/dev/drivers/scripts/stats/subseasonal

The output will be here: /lfs/h2/emc/vpppg/noscrub/perry.shafran/evs/v1.0/stats/subseasonal, in the 20240915 directories.

malloryprow commented 2 months ago

Looks good to me! Will see what @ShannonShields-NOAA says!

PerryShafran-NOAA commented 2 months ago

There are no FATAL ERRORs, but there are WARNING messages for each of the missing files, and a "Not enough files" warning to make the weekly file. It does look like the averaging of the monthly file was successful as well.

ShannonShields-NOAA commented 2 months ago

@malloryprow and @PerryShafran-NOAA Yes, everything checks out on my end too and the jobs confirmed successful with the correct WARNING behavior. This should be good for final review and merge for the new version.