Update scripts to use the GW-RT global namelist#3017
Update scripts to use the GW-RT global namelist#3017DavidHuber-NOAA merged 53 commits intoNOAA-EMC:developfrom
Conversation
dpsarmie
requested review from
WalterKolczynski-NOAA,
aerorahul and
junwang-noaa
as code owners
|
Not a fan of the hiding, at least not how it is implemented here. All the variables are still set by the parsing script, it makes the template more difficult to read, and doesn't even remove it from the rendered namelist, just comments the lines out. Given that, I'd rather just leave all the settings in the namelist with the values that turn them off. |
|
Hey Walter, Splitting up the sections into different scripts could also work, but you run into the issue of not being able to easily pass variables between scripts if they are saved as locals (correct me if I'm wrong here). Also open to your suggestions on how to best make this fit within the GW system. |
|
Here is a doc with the differences between variable values for GW and RT for a C48 atm only case: GW-RT differences We're leaning towards just changing the RT baselines to match what GW has, but we can discuss this if needed. |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
dpsarmie
force-pushed
the
feature/GW-RT_GlobalNML
branch
from
51ecf63 to
6c5b9bb
Compare
WalterKolczynski-NOAA
left a comment
WalterKolczynski-NOAA
left a comment
There was a problem hiding this comment.
Structurally it looks fine. Didn't verify that the default for every individual setting is unchanged. Just take care of the FHZER thing and I'm happy with this.
|
@dpsarmie Merge in develop again and I think this is ready for CI |
|
I was still running tests on my end as a final check but should be done before lunch. We should still talk about the final pathing to the template (just added a review for visibility). |
Added a missing period
Decrease DELTIM to 900 for C48 to accommodate SKEBINT requirement
|
Offline tests of C96C48_hybatmaerosnowDA were successful on WCOSS2. Marking ready for CI testing on Hercules. |
DavidHuber-NOAA
added
CI-Hercules-Ready
emcbot
added
CI-Hercules-Building
|
Experiment C96mx100_S2S FAILED on Hercules in Build# 2 with error logs: Follow link here to view the contents of the above file(s): (sfs_ice_prod_mem000_f024.log) |
emcbot
added
CI-Hercules-Failed
|
Experiment C96mx100_S2S FAILED on Hercules in Build# 2 in |
emcbot
removed
the
CI-Hercules-Failed
|
CI Tests set up to run in /lfs/h2/emc/ptmp/emc.global/PR/PR_3017/RUNTESTS on WCOSS |
DavidHuber-NOAA
left a comment
DavidHuber-NOAA
left a comment
There was a problem hiding this comment.
Changes look good. Just noting that I own the changes to parm/config/*/config.ufs.
|
Experiment C96C48mx500_S2SW_cyc_gfs FAILED on Hera in Build# 4 with error logs: Follow link here to view the contents of the above file(s): (enkfgfs_earc_tars_01.log) |
|
Experiment C96C48mx500_S2SW_cyc_gfs FAILED on Hera in Build# 4 in |
|
Experiment C96C48mx500_S2SW_cyc_gfs FAILED on Gaeac6 in Build# 3 with error logs: Follow link here to view the contents of the above file(s): (enkfgfs_earc_tars_01.log) |
|
Experiment C96C48mx500_S2SW_cyc_gfs FAILED on Gaeac6 in Build# 3 in |
|
Giving CI status on WCOSS2 before signing off for the weekend: testing on WCOSS2 is nearing completion with all tests completed successfully minus the extended test which is still running (with no issues so far). |
|
Experiment C96C48_hybatmaerosnowDA FAILED on Gaeac6 in Build# 3 with error logs: Follow link here to view the contents of the above file(s): (gdas_arch_tars.log) |
|
Experiment C96C48_hybatmaerosnowDA FAILED on Gaeac6 in Build# 3 in |
|
WCOSS2 CI completed minus a few gempak jobs that hit their walltimes. Still investigating cause but am setting this as passed. |
6 participants
Description
The following is a Google Doc with new variables that were added to the nml template with this PR: Google Sheets link
This PR changes the ways the input.nml file is created. Instead of creating a namelist, the variables that are set locally will now use the UFS
global_control.nml.INtemplate and the atparse function to generate the namelist.This should allow the GW branch to more easily test changes in the UFS regression test framework. It will also allow GW to more easily introduce UFS changes into GW.
The functionality in
parsing_namelists_FV3.shis mostly unchanged. Since variables are still saved aslocal, the parsing of the template will occur in this script. Sections in the namelist that were previously not generated (stoch physics options for example) will be set and parsed since the regression test template has these placeholders active. As options are turned off and on, ! will be used in the namelist to comment out sections that are not used.The MOM6 variables have been moved here since the template place holders are active in global_control.nml.IN.
Since the stoch physics options are active, the SKEB definitions were also modified in the config.efcs files. If the first value in SKEB is not -999., then the stoch physics subroutine will turn SKEB on (do_skeb set to 1).
The FHZERO variable was also renamed to reflect what was in the namelist template.
Previous PR #2733 was closed since my old branch was stale and a new branch was created.
Unresolved issues and input needed from GW group before final code review
There are some variables that were not set in the original scripts that have been added. These were set to the UFS defaults.
The tests I've conducted were to generate the namelists but more tests should be done to make sure that outputs have not changed.
Resolves #2731
Resolves ufs-community/ufs-weather-model#1664
Outstanding dependencies:
ufs-community/ufs-weather-model#2425
Type of change
Change characteristics
How has this been tested?
Tested namelist generation with C48ATM and C48S2S configurations. Further testing will need to be done to ensure intended funcitonality.
Checklist