Ufs/dev ugwp stoch phys fix

[mdtoyNOAA]

This PR fixes a bug that occurs when running stochastic physics. As part of ccpp-physics PR#22, the order in which subroutines "gwdps_run" and "drag_suite_run" were called by subroutine "unified_ugwp_run" was reversed. Subroutine "drag_suite_run" is called last, but the variable "RDXZB" (the level of the dividing streamline for blocking) is initialized, which zeros out the value that was calculated in "gwdps_run". RDXZB is needed outside of the unified_ugwp subroutine by stochastic physics. This caused stochastic physics runs to crash. The bug was not caught by regression testing.

The problem has been fixed by changing the declaration of "RDXZB" from "intent(out)" to "intent(inout)" in drag_suite.F90 and drag_suite.meta, and initializing it in drag_suite.F90 only when the GSL drag suite blocking is performed, i.e., when "do_gsl_drag_ls_bl" is equal to .true.

The stochastic physics team (@pjpegion et al) has tested the revised code, and the crashes no longer occur. Regression tests and ORT's have passed (Hera -- Intel) with no changes to the baseline.

This PR fixes Issue #62.

This fix needs to be done for running the GEFS.v13 reanalysis and reforecasts.

[yangfanglin]

@grantfirl @dustinswales @Qingfu-Liu Please consider giving this PR the highest priority for committing. It is needed for running the Ensemble Protype-4 for GEFS.v13 development.

@junwang-noaa A new model tag, for instance, GFSv17.HR1b, needs to be created after this PR is committed to the latest develop branch. -- Thanks.

[junwang-noaa]

@yangfanglin Do you mean we will create a new cut HR1 tag from the latest develop branch after this PR is committed? This will introduce 2 months of model code updates since Jan 30 when the previous HR1 tag was created. Also global workflow is using current HR1 tag, it may break before the code changes in the last two months are brought to global workflow.

[Qingfu-Liu]

@fanglin Yang - NOAA Federal ***@***.***> Question about this PR: do you want only to fix the bug for the HR1 (which is tagged in late January), or use this new PR which includes two month's development for the UFS model and physics?

…

On Wed, Apr 5, 2023 at 11:50 AM Grant Firl ***@***.***> wrote: ***@***.**** approved this pull request. — Reply to this email directly, view it on GitHub <#61 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGTS6URIUVERXABLXWOG7L3W7WIFFANCNFSM6AAAAAAWTQUWXE> . You are receiving this because you were assigned.Message ID: ***@***.***>

[lisa-bengtsson]

I don't think we should make a model tag HR1b including the two months of model code changes. Why not include it in the ensemble prototype runs, and have it as an update for HR2?

[yangfanglin]

@junwang-noaa I do not think there are any recent model changes that will break the ensemble forecast. But you have a good point about the workflow. My understanding is that the team is not, and will not use the official workflow for conducting EP4 and reforecast experiments. WalterKolczynski-NOAA and @bingfu-NOAA can you confirm ?

[yangfanglin]

There are two options. (1) commit this PR to the development branch since it is a bug fix, create a new tag HR1b for the ensemble team. (2) commit this bug fix to both the development branch and update the HR1 tag.

My preference is (1) since there are some recent infrastructure related changes that can potentially improve computational performance.

For example, using the newly added capability of "Write restart files using the write grid component" may lead to reduced CPU and node cost for conducting the massive reforecast experiments.

[bingfu-NOAA]

@yangfanglin @junwang-noaa
Current global-workflow can not be used to run ensemble experiments, all the ensemble prototypes used modified global-worklow (adding specific ensemble settings and stochastic settings) for experiments.

[Qingfu-Liu]

@jun Wang - NOAA Federal ***@***.***> As Fanglin suggested, this PR #61 (which includes the two month's update for UFS models) can be merged into the repository and created a new HR1b for the ensemble team. Who is going to do this?

…

On Wed, Apr 5, 2023 at 12:23 PM Bing Fu ***@***.***> wrote: @yangfanglin <https://github.com/yangfanglin> @junwang-noaa <https://github.com/junwang-noaa> Current global-workflow can not be used to run ensemble experiments, all the ensemble prototypes used modified global-worklow (adding specific ensemble settings and stochastic settings) for experiments. — Reply to this email directly, view it on GitHub <#61 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGTS6UQVZTZOZTYPEF5ERV3W7WL7VANCNFSM6AAAAAAWTQUWXE> . You are receiving this because you were assigned.Message ID: ***@***.***>

[junwang-noaa]

@DeniseWorthen FYI.

[DeniseWorthen]

@Qingfu-Liu I will create a new HR1b tag when PR #61 is merged. It will be created from the current develop branch.

[bingfu-NOAA]

@junwang-noaa @yangfanglin I think option (1) as suggested by Fanglin is also preferable to ensemble experiments.

[lisa-bengtsson]

@bingfu-NOAA PR #57 is a bug fix as well that you may want to include in your prototype runs if they have not started yet. It should be merged next in line.

@yangfanglin maybe we should carefully look at all the updates in all the model components between HR1 and HR1b? Do we know if there is anything that requires namelist updates in the ensemble workflow in the other component models? There were quite many PR's merged in the last 2 months.

[yangfanglin]

@lisa-bengtsson HR1 tag was created after this commit
Bug fix for cloud effective radius for convective clouds (HR1) (https://github.com/ufs-community/ufs-weather-model/pull/1582[)](https://github.com/ufs-community/ufs-weather-model/commit/c22aaadfb8dafb5687e2f3bf6416e8f11e3f9a1d)

One can find at https://github.com/ufs-community/ufs-weather-model/commits/develop all commits after the above PR.

[bingfu-NOAA]

@lisa-bengtsson Thank you for pointing it out. Yes, ensemble prototype 4 has not started yet. I think we can use that bug fix once it is committed.

[lisa-bengtsson]

@lisa-bengtsson HR1 tag was created after this commit Bug fix for cloud effective radius for convective clouds (HR1) ([https://github.com/ufs-community/ufs-weather-model/pull/1582)

One can find at https://github.com/ufs-community/ufs-weather-model/commits/develop all commits after the above PR.

Thank you @yangfanglin I will take a look. I don't really know exactly which PR's from waves, ice and ocean that will actually be used in our prototype runs from this list, but we could maybe bring it up in the next meeting on prototypes.

[junwang-noaa]

@lisa-bengtsson @yangfanglin @bingfu-NOAA I don't see updates from ocean, ice, wave that will change model results except PR#1692, which is expected to improve the atm-wave coupling. But we do have the restart filename updates for atm/ocn/ice/wave, which may require corresponding updates. The rest are physics updates. @DeniseWorthen FYI.

[pjpegion]

How about tagging this as EP4?

[lisa-bengtsson]

@junwang-noaa thank you! It looks like no other physics updates besides #57 and #61 will impact GEFS/GFS.

[DeniseWorthen]

@yangfanglin Was there a final decision on what to tag the new UFS hash once this is fully merged? I see a suggestion of EP4 instead of HR1b.

[yangfanglin]

@DeniseWorthen Yes, EP4 is the favorite by consensus.

[DeniseWorthen]

Tag EP4 has been created.

[dustinswales]

Corresponding EP4 tag created in ccpp-physics

[bingfu-NOAA]

Thank you all for helping on making the PR and creating the tag in a short time.