Update calculation of time-averaged radiation variables#900
Update calculation of time-averaged radiation variables#900LarissaReames wants to merge 19 commits into
Conversation
…te at sub-radiation-timestep output frequency.
…t.F90 in to GFS_radiation_post.F90 in ccpp/physics, requiring changes to all suites.
|
@LarissaReames-NOAA Is there still work going on in this PR, or is there another reason that it is in draft mode? Also, I'm not seeing an associated ufs-weather-model PR. |
I think we're done with all of the ccpp-physics reviews at this point? I can go ahead and move from draft to active. I guess I wasn't sure if I needed a UWM PR if no changes were required in that repo for this PR to work. I've never submitted a PR that didn't have at least a change or two at the UWM level. |
Yes, ufs-community/ccpp-physics#241 has all the approvals it needs. It's OK to have a UWM PR with only an update to submodules, fv3atm and ccpp-physics in this case. This happens all the time when there is a physics-only change. UFS code managers still need something to "attach" submodule changes to in order to test it. Plus, when you run regression tests on whatever platform that you have access to, you need to upload your "pre-test" RT log and the test_changes.list file, so the UWM PR is never completely empty even with only submodule changes. |
Hah, total brain fart on that one. Of course I'll have to run the regression tests again. Okay, I'll open a UWM PR. |
… bug fix and workaround for bad tsfco NOAA-EMC#916 (NOAA-EMC#896) * Parallel IO enhancements from GDIT * Add SW clear-sky downward flux at surface to available diagnostics * fix compilation error in GFS_diagnostics.F90 * update atmos_cubed_sphere to turn ENABLE_PARALLELRESTART OFF by default to avoid RT errors on some platforms (Hera, Hercules at least) * ccpp-physics - wet() bug fix and workaround for bad tsfco --------- Co-authored-by: Dustin Swales <dustin.swales@noaa.gov> Co-authored-by: Samuel Trahan <samuel,trahan@noaa.gov>
|
@DusanJovic-NOAA This can be closed since it has been replaced by #941 since @LarissaReames-NOAA was a probationary fire. |
3 participants
Description
This PR fixes incorrect diagnostic output for fhzero-averaged radiation output variables. Presently, there is a significant "sawtooth" appearance to these fields (such as dswrf_ave and ulwrf_avetoa) when we expect smoother lines with jumps only at fhzero diagnostic reset times. Here is an example of the present (OLD) result and the result from the current PR (FIX) from a 24-hour S2S regression test forecast.
Issue(s) addressed
Testing
Ran full regression test suite on Hera. See log files here:
/scratch1/NAGAPE/hpc-wof1/lreames/ufs-weather-model/tests/logs/log_heraForecast output can be found here:
/scratch1/NCEPDEV/stmp2/Larissa.Reames/FV3_RT/rt_4109245All regression tests fail as sfcf*.nc files, as well as post-processed grib2 files as a result, are not identical. New baselines will be necessary.
Dependencies
Requirements before merging