NCAR/main ccpp-physics sync#2998
Conversation
|
@grantfirl I tried to clone your repo to run the mpas test on Hercules, and it's giving me an error cloning https://github.com/grantfirl/ufsatm, I think because your repo is still labeled fv3atm, not ufsatm. Could you either update the repo or update the .gitmodules? When I click from your WM repo to UFSATM@3ba0301 I get a 404 Not Found (presumably for the same reason). |
OK, I renamed my repo to ufsatm, so it should be OK now. Please let me know. I also updated this PR set to the latest develop. |
|
Looks good, thanks! :) |
|
@grantfirl Could you sync w/develop? We are ready to begin processing this PR. |
@gspetro-NOAA Finished syncing with develop. |
|
Acorn done. WCOSS2 due to prod switch means we had to sync over the latest develop- and input-data and that's been taking a while. I'm going to let all the syncing finish overnight and get the baselines created and the full suite run tomorrow. |
|
@gspetro-NOAA done, finally! |
|
@grantfirl new ufsatm hash is NOAA-EMC/ufsatm@3aba288 |
Commit Queue Requirements:
test_changes.listindicates which tests, if any, are changed by this PR. Committest_changes.list, even if it is empty.Description:
This is primarily a ccpp-physics sync from NCAR/main back to the ufs-community/ufs/dev branch. Some changes to UFSATM files were necessary are as follows:
xr_conandxr_epsnamelist variables to allow recent changes to the Xu-Randall cloud fraction algorithms configurable by the host. If these variables are not set in the namelist (which is the default for all current workflows and UFS RTs), these are set using the same logic that was used in radiation_clouds.Fnumber_of_diagnostic_bucketsvariable and array dimension growth for 2 diagnostic variables. By default this is set to 1 and the change is transparent for UFS users but this functionality could be used in the future if multiple bucket-clearing intervals are desiredCommit Message:
Priority:
Git Tracking
UFSWM:
Sub component Pull Requests:
UFSWM Blocking Dependencies:
Documentation:
Changes
Regression Test Changes (Please commit test_changes.list):
All of the baseline changes are a result of removing two openmp statements in radiation_aerosols.F within the CCPP physics. See https://github.com/ufs-community/ccpp-physics/pull/328/files#diff-17d80e12dfd7aa6ad66670084f0c3e002d18db4c3d7af57df61886c3a1547eaa for the changes leading to the results change. It was also found that the hafs_regional_atm_thompson_gfdlsf_intel test changed the baselines due to one of the physics changes from NRL. The precise code change culprit wasn't pinned down and it wasn't obvious from the original ccpp-physics PR.
Input data Changes:
Library Changes/Upgrades:
Testing Log: