Emc post ssmis f17#367
Conversation
…ew.text, remove the params_grib2_tbl_new_ssmisf17
|
@LinZhu-NOAA Can you also sync your branch with upstream/develop with the instructions at |
|
@WenMeng-NOAA The featured branch ems_post_ssmis-f17 is synced with upstream/develop. |
|
@LinZhu-NOAA Thanks for the update. |
|
@LinZhu-NOAA To generate new dataset HURSAT, the fix files from crtm module are needed. Can you provide me the configuration of your test so I can update the baseline of hafs in the UPP rt tests? |
|
@WenMeng-NOAA I ran the tests on Luna. I modified the scripts for HAFS regression tests as follows: /gpfs/hps3/emc/hwrf/noscrub/Lin.L.Zhu/save/EMC_post_RT/jobs-dev/run_post_fv3hafs_WCOSS_C.sh eg. run_post_fv3hafs_WCOSS_C.sh Thanks, |
|
@LinZhu-NOAA I conducted the UPP standalone tests for hafs. Can you verify the grib2 data sets HURPRS09.tm00 and HURSAT09.tm00 at /u/Wen.Meng/ptmp/fv3hafs_2020082512 on mars? Thanks! |
|
@WenMeng-NOAA There is only HURPRS09.tm00 in /u/Wen.Meng/ptmp/fv3hafs_2020082512/ on Mars. I also check the /u/Wen.Meng/ptmp/fv3hafs_2020082512/postxconfig-NT.txt, there is no satellite products in this configure file, it seems old hafs configure file. Thanks! |
|
@LinZhu-NOAA Can you double check the directory /u/Wen.Meng/ptmp/fv3hafs_2020082512 from WCOSS-Dell Mars, not luna? The actual path is /gpfs/dell2/ptmp/Wen.Meng/fv3hafs_2020082512. Thanks! |
|
@WenMeng-NOAA Sorry, I checked the files in wrong directory. Yes, these output looks good to me. Thanks! |
|
@LinZhu-NOAA Thanks for confirming. I will proceed the final testing for merging. The UPP regression test baseline for hafs will be recreated with this PR merged. |
|
@WenMeng-NOAA Thanks a lot! BTW, will you generate the new lib for in-line post after merging? Thanks! |
|
@LinZhu-NOAA The new upp lib will be generated in September. The NCEPLIBs team would like the libs at the hpc-stack not upgraded too frequently. |
|
@WenMeng-NOAA Okay, got it. Thank you very much! |
|
The UPP regression tests were conducted on WCOSS-Dell, WCOSS-Cray and Hera. With this PR merged, the baseline of hafs will be re-created. This PR is ready for merging. |
orog_files_mdl="${CRES}${DOT_OR_USCORE}oro_data.tile${TILE_RGNL}.halo${NH4}.nc"
to run workflow with the UFS_UTILS GNU build.
* Update order of variable definition files in ccpp/config/ccpp_prebuild_config.py * Remove duplicate variables in ccpp/data/GFS_typedefs.meta
2 participants
Add the SSIMS-F17 channels to grib2 output for HAFS development. The key modifications are as follows:
a) organized non-satellite products and add 8 satellite products including 4 goes-r and 4 ssmis-f17 channels
postcntrl_hafs.xml
postxconfig-NT-hafs.txt
b) add ssmis-f17 channels
post_avblflds.xml
arams_grib2_tbl_new
a) RQSTFLD.F
Line 2367-2379: add ssmis-f17 channels (825 - 828)
b) CALRAD_WCLOUD_newcrtm.f
Line 208: add local logical variable post_ssmis17
Line 290-294: set post_ssmis17
Line 334: add post_ssmis17 to the list
Line 456-469: set ssmis17 channels (channel 15-18)
Line 485-487: take out the ssmis_f17 channel selection for grib1
Line 542: add ssmis_f17
Line 1249: add ssmis_f17
Line 1808-1825: modify output block for ssmis_f17
For hafs regression tests, it need to link the crtm fix files to run directory since satellite products are added to grib2 control file.
export FIXcrtm=/gpfs/dell1/nco/ops/nwprod/lib/crtm/v2.3.0/fix
/gpfs/hps3/emc/hwrf/noscrub/Lin.L.Zhu/save/hafs_20210527/ush/link_crtm_fix.sh ${FIXcrtm}