Modifications for gfsda.v16.3.0 branch

[emilyhcliu]

Change the following to gfsda.v16.3.0 branch:
A. /scripts/exglobal_atmos_analysis.sh:
(1) set cao_check to true
(2) set ta2tb to true
(3) set rmesh(4)=50 for AVHRR and VIIRS (NSST application)
(4) point CRTM cloud coefficient file to GFDL one

B. /modulefiles/gsi_common.lua:
(1) change crtm_ver from 2.3.0 to 2.4.0

C. Create specific fix branch which include fix files for global application only (DA_GFSv16.3.0_global_only)

[emilyhcliu]

Tagging @ADCollard @HaixiaLiu-NOAA @XuLi-NOAA
Please take a look at these changes. Thanks.

[HaixiaLiu-NOAA]

changes look good to me.

[emilyhcliu]

@MichaelLueken-NOAA We have some changes in GSI fix files.
We created a subdirectory under /GSI/fix/gfsv16_historical/gfsv16.3

There are two historical satinfo files and two historical convinfo files.
They are specifically updated for the gfsv16.3 paralle testing (the parallel experiment starts from 20211015 18z)

Could you help us add this new directory into the gfsv16_historical?
Thanks a lot.

[RussTreadon-NOAA]

Both release/gfsda.v16.3.0 and release/gfsda.v16.3.0_emily currently point at fix @ 99f147c. This is not correct for implementation. fix @ 99f147c contains more than just global GSI fix files. NCO only wants global GSI fix files in the implementation package.

For example, the fix associated with WCOSS2 tag gfsda.v16.2.0 is DA_GFSv16.1.8_global_only We need to create a trimmed down (global only) fix for the GFS v16.3 implementation.

[emilyhcliu]

@ADCollard and @HaixiaLiu-NOAA
I created a subdirectory under /GSI/fix/gfsv16_historical/gfsv16.3

There are two historical satinfo files and two historical convinfo files.
They are specifically updated for the gfsv16.3 paralle testing (the parallel experiment starts from 20211015 18z)

The global-workflow will these the convinfo and satinfo according to the cycle date/time from this directory.

Please see the corresponding PR in global-workflow related to this PR
global-workflow PR #876

[emilyhcliu]

Both release/gfsda.v16.3.0 and release/gfsda.v16.3.0_emily currently point at fix @ 99f147c. This is not correct for implementation. fix @ 99f147c contains more than just global GSI fix files. NCO only wants global GSI fix files in the implementation package.

For example, the fix associated with WCOSS2 tag gfsda.v16.2.0 is DA_GFSv16.1.8_global_only We need to create a trimmed down (global only) fix for the GFS v16.3 implementation.

@RussTreadon-NOAA Thanks for pointing out the issue in the fix file. I will use the gfsda.v16.2.0 as an example to create a proper GSI fix for v16.3 implementation.

[MichaelLueken]

@emilyhcliu Fix submodule branch DA_GFSv16.3.0_global_only was created from rev2 @ 99f147c. To add the gfsv16.3 subdirectory to DA_GFSv16.3.0_global_only and update the fix submodule revision in release/gfsda.v16.3.0 -

In release/gfsda.v16.3.0_emily:

• git remote update
• cd fix
• git checkout DA_GFSv16.3.0_global_only
• Add the gfsv16.3 subdirectory to gfsv16_historical.
• git add gfsv16_historical/gfsv16.3
• git commit
• Remove fix files not used with global. I recommend the above commit in case you make a mistake while removing fix files, you can back out (git reset --hard HEAD~1) without removing the new subdirectory.
• git commit
• git push origin DA_GFSv16.3.0_global_only
• cd ..
• git add fix
• git commit
• git push origin release/gfsda.v16.3.0_emily

Once the changes in this PR are merged to release/gfsda.v16.3.0, the fix submodule revision will be updated to DA_GFSv16.3.0_global_only.

[emilyhcliu]

@MichaelLueken-NOAA Thanks for your detailed instruction on how to create the global only fix directory under gfsda.v16.3.0
I had followed your instruction and update the release/gfsda.v16.3.0 fix with the DA_GFSv16.3.0_global_only.

Tagging @HaixiaLiu-NOAA @ADCollard @RussTreadon-NOAA

[emilyhcliu]

@HaixiaLiu-NOAA @ADCollard @MichaelLueken-NOAA @RussTreadon-NOAA
If there is no objection to the changes made in this PR:
fix
modulefiles/gsi_common.lua
scripts/exglobal_atmos_analysis.sh

I will merge this branch back to release/gfsda.v16.3.0

Please give me an OK or let me know if you have any questions.
Thanks.

[ADCollard]

Looks good to me

[MichaelLueken]

@MichaelLueken-NOAA Thanks for your detailed instruction on how to create the global only fix directory under gfsda.v16.3.0 I had followed your instruction and update the release/gfsda.v16.3.0 fix with the DA_GFSv16.3.0_global_only.

@emilyhcliu I just checked the changed files in this PR and I don't see the update to fix. I do see that the DA_GFSv16.3.0_global_only submodule branch has had the gfsv16.3 subdirectory added (@ 00fe93) and the fixed files were pruned for global only (@ 9eed37), but it doesn't seem as though these modifications made it back into release/gfsda.v16.3.0_emily. Please check if you haven't pushed your release/gfsda.v16.3.0_emily branch with the updated fix files as well.

[emilyhcliu]

@MichaelLueken-NOAA I did the following after I pruned the files in the fix directory:

git push origin DA_GFSv16.3.0_global_only
cd ..
git add fix
git commit
git push origin release/gfsda.v16.3.0_emily

[XuLi-NOAA]

Emily, rmesh(4)=50.0 is the final decision? There seems to be no final decision yet. Xu

…

On Thu, Jun 16, 2022 at 2:44 PM emilyhcliu ***@***.***> wrote: @emilyhcliu <https://github.com/emilyhcliu> requested your review on: #412 <#412> Modifications for gfsda.v16.3.0 branch. — Reply to this email directly, view it on GitHub <#412 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANEMLJT2AXO24QSIQTYDUWLVPNYWVANCNFSM5Y7W2PPQ> . You are receiving this because your review was requested.Message ID: ***@***.***>