Add Tom's regional DA dycore fix

[junwang-noaa]

1. delz halo fix
2. changes for putting GSI anl into boundary file in a more robust manner in SAR DA system

Related PRs:
fv3atm #132
ufs-weather-model #154

[climbfuji]

Is this specifically tailored to GFDL-MP? How about other microphysics schemes that use different tracers?

[EricRogers-NOAA]

No, it has been run for Thompson MP as well as GFDL MP.

[climbfuji]

Ok, thanks.

[danholdaway]

Thanks for including my changes Jun. Not sure where this happened and no problem if it doesn't make sense to include in this PR since this file wasn't touched and I can make another one but theres an additional change for when not using the INTERNAL_NML_FILE definition. In the file model/multi_gases.F90 at line 42 it should be use fms_mod, only: check_nml_error, open_namelist_file. We'll work to get away from using this compiler definition since it's not recommended/supported but we could do with keeping it functional for now. I'll approve anyway in case you want to leave it out of this PR.

[danholdaway]

Thanks for including my changes Jun. Not sure where this happened and no problem if it doesn't make sense to include in this PR since this file wasn't touched and I can make another one but theres an additional change for when not using the INTERNAL_NML_FILE definition. In the file model/multi_gases.F90 at line 42 it should be use fms_mod, only: check_nml_error, open_namelist_file. We'll work to get away from using this compiler definition since it's not recommended/supported but we could do with keeping it functional for now. I'll approve anyway in case you want to leave it out of this PR.

Apologies, copy paste error. It should be:
use fms_mod, only: check_nml_error, open_namelist_file, close_file
on that line

[junwang-noaa]

I see. We've about to commit for Tom's fix. I'd suggest to create another dycore PR, we will get it in in next commit.

…

On Tue, Jul 14, 2020 at 12:30 PM Daniel Holdaway ***@***.***> wrote: Thanks for including my changes Jun. Not sure where this happened and no problem if it doesn't make sense to include in this PR since this file wasn't touched and I can make another one but theres an additional change for when not using the INTERNAL_NML_FILE definition. In the file model/multi_gases.F90 at line 42 it should be use fms_mod, only: check_nml_error, open_namelist_file. We'll work to get away from using this compiler definition since it's not recommended/supported but we could do with keeping it functional for now. I'll approve anyway in case you want to leave it out of this PR. Apologies, copy paste error. It should be: use fms_mod, only: check_nml_error, open_namelist_file, close_file on that line — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#26 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AI7D6TN2Y4JAV4CIJD254RDR3SBYZANCNFSM4OHWZUMA> .