global_ens: SST ice fix #668
Conversation
|
I see that there are no changes needed for the plots. global_det and subseasonal needed changes in the plots step to be able to pick up the new threshold. I know that global_ens uses different plotting scripts than global_det and subseasonal, but just I wanted to check. |
1. jevs_global_ens_gefs_atmos_sst_statsLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr668/EVS/dev/drivers/scripts/stats/global_ens/jevs_global_ens_gefs_atmos_sst_stats.o181041603 |
The global_det and subseasonal components use Grid_Stat and cnt_thresh and cnt_logic config options to filter out data < 271.15 K, and the threshold values appear in the FCST_THRESH and OBS_THRESH columns of stat files and plot titles. So, they need to adjust the plotting code to remove the threshold notation in the plot titles. The global_ens component uses Ensemble_Stat and censor_thresh and censor_val config options, and the threshold values do not appear in the FCST_THRESH and OBS_THRESH columns. So, no impacts to the plot titles and plotting code. I set |
The output looks good to me. |
|
Noted, thanks for the explanation! |
There was a problem hiding this comment.
Code changes are good and testing successful.
Thanks for these changes @GwenChen-NOAA!
|
Just updated para to bring in these changes. jevs_global_ens_gefs_atmos_sst_stats already ran for the day but so the changes will be in the stats starting tomorrow for the parallel. |
|
Is it possible to bring these changes to prod via a EVS bugfix? The global_ens SST verification shown in the prod webpage is way off. |
|
@GwenChen-NOAA Unfortunately, it is not possible to submit another EVS bugfix code delivery to NCO at this time. We've already sent them two and they weren't supposed to accept any from us! I will break the SST link on the webpage to prevent the prod images from being displayed. The SST fixes for global_ens, global_det, and subseasonal will need to make their way to prod in EVS v2.0. Thanks for asking! CC @malloryprow |
|
Ok, I will take down the global_ens SST prod webpage and start archiving sst stat files from emc.vpppg parallel to cut down the days needed for rerun. cc @ShannonShields-NOAA and @QiShi-NOAA for their awareness. |
|
@GwenChen-NOAA Thanks! I updated the global_ens webpage this morning to prevent the incorrect SST images from displaying on the /prod webpage. I think this quick fix would be much less work than you removing the SST option from the left-hand side menus on all of the global_ens pages, just for you to add it back again later, etc. I hope this solution works for you! |
|
Thanks, @AliciaBentley-NOAA! I saw the image section is blank now. |
Description of Changes
This PR fixes the SST issue with ice temp definition discrepancies between the forecasts and observations by filtering out data < 271.15 (K) in Ensemble _Stat. Test run results are available at
https://www.emc.ncep.noaa.gov/users/verification/global/gefs/test/atmos/grid2grid/sst/
Developer Questions and Checklist
${USER}where necessary throughout the code.HOMEevsare removed from the code.dev/drivers/scriptsordev/modulefileshave been made in the correspondingecf/scriptsandecf/defs/evs-nco.def?Testing Instructions
(1) Set up jobs
a. Symlink the EVS_fix directory locally as "fix".
b. In the driver scripts, edit the following environment variables:
HOMEevs - set to your test EVS directory
COMIN - set to /lfs/h2/emc/vpppg/noscrub/emc.vpppg/${NET}/$evs_ver_2d
COMOUT - set to your test output directory
(2) Run jobs
Run the following job in EVS/dev/drivers/scripts/stats/global_ens:
qsub jevs_global_ens_gefs_atmos_sst_stats.sh