iovr=4 cleanup in RRTMG #780
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
climbfuji
reviewed
Nov 16, 2021
|
icloud=3 was corresponding to Thompson cloud scheme, which was implemented
to work with the Ferrier-Aligo MP scheme as is in the 2018 operational
HWRF. However, I see that there is a PR coming in to update cal_cldfra3,
which is still in radiation_clouds.f. All WRF-favored subroutines related
to iovr=4 are obsolete and deleted as requested.
…On Tue, Nov 16, 2021 at 2:32 PM Dom Heinzeller ***@***.***> wrote:
***@***.**** commented on this pull request.
I see that you are removing the icloud=3 and iovr=4 paths completely. Are
there any additional changes needed so that iovr option 4 follows the
same path as the other iovr options?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#780 (review)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AG7TW2UPK5LDEZGDZBIT3BLUMLEVVANCNFSM5IFJDLRQ>
.
|
|
@climbfuji I took a quick look at the changes in PR #780, and I spotted a couple of things that I recommend changing.
|
|
Thanks for your feedback. I will follow up with you when it is ready.
…On Wed, Nov 17, 2021 at 1:33 PM Mike Iacono ***@***.***> wrote:
@climbfuji <https://github.com/climbfuji> I took a quick look at the
changes in PR #780 <#780>, and I
spotted a couple of things that I recommend changing.
1. First, in radlw_main.F90 line numbers 912 and 1042 are identified
for removal, but these lines must be retained to convert alpha from a 2-D
array to the 1-D array alph for later use in the code. In radsw_main.F90
the comparable line numbers 1043 and 1156 must also be retained.
2. Also, please retain the entire subroutine cldprmc in radlw_main.F90
for now, since it may be needed to address item 3 in issue #784.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#780 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AG7TW2X43RX3BQAO2SNTAJDUMQGRHANCNFSM5IFJDLRQ>
.
|
gthompsnWRF
reviewed
Nov 23, 2021
…rt_iovr4 * 'main' of https://github.com/NCAR/ccpp-physics: Add -O1 for RRTMGP file physics/rte-rrtmgp/rrtmgp/kernels/mo_gas_optics_kernels.F90 back in Remove Laurie from CODEOWNERS permit few max ice number conc; scale back cloud fraction a little; reduce assumed snow as ice in IWP for radiation Fixed error in rte-rrtmgp submodule hash. Updated rte-rrtmgp submodule. Performance improvements. Update physics/GFS_debug.F90 with additional changes from recent commits Update physics/GFS_debug.F90 with changes to surface emissivity variables fix the vfrac over glacier Fixed typo. fix snow depth issue in Noah MP limit pahi to Noah MP only fix the typo address some comments from reviewers address some comments from reviewer Update sfc_noahmp_drv.F90 Update GFS_surface_generic.F90 Update module_sf_noahmp_glacier.f90 Update sfc_noahmp_drv.F90 Update sfc_diag_post.F90 Update module_sf_noahmp_glacier.f90 Update GFS_surface_loop_control.F90 Update GFS_surface_generic.F90 update rte-rrtmgp repository Update GFS_surface_generic.meta Update GFS_surface_generic.F90 Update GFS_surface_loop_control.meta Update GFS_surface_loop_control.meta minor rearrangement of cloud fraction with very low mixing ratios disallow rain in Thompson to be used in radiation scheme; also set fractional cloudy LWP and IWP inside progcld6 per review by climbfuji, make suggested changes Add 'in Pa' to three long names to clarify the variables re-use icloud=3 option for progcld_thompson tiny adjustment of indent/format consistency per discussions with Ruiyu, increasing min snow and graupel size seems advantageous fix a few comment lines per review of pull request 781 silly stray comment/exclamation point Change units of relative humidity variables in CCPP metadata from 1 or none to frac Revert unnecessary whitespace changes in physics/module_mp_thompson.F90 Use correct vertical dimension for several cloud arrays in physics/GFS_rrtmg_post.meta and physics/GFS_rrtmg_pre.meta minor adjustments to reduce coverage of partly cloudy conditions Move total albedo to right place in physics/GFS_rrtmg_post.meta, make intent(inout) Avoid dividing by zero in physics/GFS_rrtmg_post.F90 Fix compilation error with gfortran in physics/radiation_clouds.f remove trailing whitespace Add @mkavulich to CODEOWNERS update the namelist variables to control semi-lagrangian sedimentation bug fix lwp_ex and iwp_ex inside progcld6; factor of 1000 Some housekeeping make icloud=3 call cldfra3 compatible with updated subroutine Updated rte-rrtmgp submodule. Small change to accommodate in.meta files in ccpp-physics. correct a dumb mistake added a new cloud fraction scheme (actually upgraded cal_cldfra3) designed with Thompson-MP for GFSv17-prototype8 adding a debug option for reading a new month data put a few lines of code in read_netfaer Modified 2 ugwp metadata files to fix ufs-community/ufs-weather-model 'fhzero reproducibility' issue NCAR#908 Fix issue with surface-albedo in RRTMGP. Fix aerosol diagnostics. Fix omission from conflict. Expanded format fields for percentages to allow 100% to be output. Meta data fix. Bug fix for decomposition tests. add a output information for reading a new month merra2 handle a case with n2=13 delete some debug stuff Replace units 'various' with 'mixed', update several invalid units of non-physical quantities, fix units of humidity diagnostic variables decreases memory usage by 6 times for MERRA2 to address reviews add semi-Lagrangian sedimentation of graupel into Thompson MP Some cleanup for P8. Update rte-rrtmgp remove some false values over the glacial ice correct the inout issue of smc in sflx.f pass the right zvfun from noahmp to PBL Adding GFS surface scheme to Noah MP as one of opt_sfc options remove the iteration loop for Noah MP and implement a new coupling interface between atmos. and Noah MP modification to add extra land fields to the output modification to add extra land fields to the output add Moorthis gcycle fix for tiled fixed file
|
@mjiacono I made changes as you requested. |
|
@mzhangw The changes you made in response to my requests look OK to me. |
…rt_iovr4 * 'main' of https://github.com/NCAR/ccpp-physics: Remove a non-existing file from list of SCHEMES_OPENMP_OFF, compile mo_gas_optic_kernels.F90 with -O1 in PROD/REPRO mode Update DDTs in physics/GFS_debug.F90 Address reviewer comments Remove variables that no longer exist from physics/GFS_debug.F90 Bug fix in the computation of soil moisture conductivity. The bug was causing cold/moist bias. Update minimum cmake version to remove legacy code, reduce chatter Major cleanup of cmake build config Remove additional newline in CODEOWNERS Fix wrong dimensions in physics/drag_suite.meta Update GFS_debug.F90, add GFS_checktracers debugging routine Adjust format statement in physics/mp_thompson.F90 Fixed sign error in SSGWD component of drag_suite.F90 1. Pass in flag_iter to MYNNSFC_wrapper for consistency with iterations in LSM, ice and ocean. ! when iter = 1, flag_iter = .true. for all grids ! when iter = 2, flag_iter = .true. when wind < 2 for both land and ocean (when nstf_name1 > 0) 2.Added flag_iter to all computations in MYNN surface layer scheme to avoid inconsistencies with iterations in LSM, ice andocean models. 3.Removed averaging: 0.5*(tsurf + tskin) and used tskin instead, because tsurf could be not defined. 4.Added checks on QSFC for non-defined values in LSM, ice and ocean models. If there are not defined values, compute QSFC as at itimestep=1. Move PSFC computation to the beginning before it is used. Added computations of local TSK_wat, TSK_ice and TSK_lnd at ITIMESTEP=1 before they get used. Fixed the bug in horizontal dimension for xlon_d. Restored modified by accident print statement. Fixed bugs in MYNN surface layer scheme in land/ice/water parts of the code to avoid using in computations "huge" numbers of qsfc_lnd, qsfc_ice and qsfc_wat. Also, fixed bugs in MYNN surface layer scheme for initializattion and consistent use of surface QV from RUC LSM over land and ice. Pass into the RUC driver lat/lons for debugging. Bug fix in physics/GFS_rrtmg_pre.meta, use correct vertical dimension Fix bugs in sfc_drv_ruc.F90 introduced by the merge process bug fix: must initialize dth prior to moisture_check routine Updates to MYNN-EDMF Turn on GF aerosol-awareness Tune clwdet (cloud-water detrainment) Make evfact (evaporation factor) and radiation tuning factors scale-aware Update to physics/cu_gf_driver.meta following CCPP standard names update Fix intent of variable qci_conv in physics/GFS_rrtmg_pre.meta Bug fixes for uninitialized variables in physics/cu_gf_deep.F90 and physics/cu_gf_driver.F90 Update physics/GFS_debug.F90 with latest changes to GFS_interstitial DDT Update standard names mid_conv_activity_counter and gf_mid_memory_counter Clean up code Fix definition of xland1 Clean up of code syntax Slight cleanup to kklev 1. Changes for fracional grid (frac_grid=.true.) 2. Replace logical variable for lakes from 'lake' to 'use_flake' 3. Added spp code from WRF version of RUC LSM in case it would be needed for RRFS. So far it is not hooked up to SPP weights. Fix for uninitialized kklev for mid clouds Fix cactiv_m error Update to the entrainment formula - allows removal of contraint on accelerating plumes. MYNN-EDMF updates and bug fixes Fix component issue with xland Updates to prepare for merger with latest version of master Code clean up in GFS_rrtmg_pre.F90 Code clean up for GFS_rrtmg_pre.F90 Remove an extra space - Turn off aerosols in GF - Improve GF-radiation coupling when GF substituted for SAS in the GFS physics suite Fix b4b differences for GSD v0 (RUC LSM, tiice) Revert change to CODEOWNERS Add GF variables to physics/GFS_debug.F90 Remove trailing whitespaces in physics/cu_gf_driver.F90; bug fixes in physics/cu_gf_driver.F90 to achieve b4b identical results in restart runs Correct wrong standard names in physics/cu_gf_driver_post.meta and physics/cu_gf_driver_pre.meta Taking out OPTIONAL declaration, as it was before. Modified aod_gf and cactiv_m so that work appriopriately with restart Minor code clean up Removing some commented out code. Another bugfix in physics/cu_gf_deep.F90 to prevent qc(i,k) being zero Bug fix for clw_allh Remove unneeded variable fix spelling more doxygen complaints (I think), but they dont make sense. Testing changes by trial and error... more doxygen complaints (I think), but they dont make sense. Testing changes by trial and error... Bug fixes - complaining about comments again... Bug fixes - removing a 3rd doxygen bug Bug fixes - removing a 2nd doxygen bug Bug fixes/clean up: (1) removing doxygen bug, (2) removing j indices from the driver. Bug fix Update of GF aerosol treatment and tunings Final updates for the MYNN-EDMF for HFIP Updating MYNN-EDMF
Merged
13 tasks
grantfirl
reviewed
Dec 16, 2021
|
Other than my comment about cleaning up now-unused modules, these changes look OK to me. Since they address an issue raised by @mjiacono and he has approved, then it is good enough for me. @mzhangw Do you know if the removed code had been used in any of the UFS regression tests? I need to know whether these changes will require new baselines or not. |
There are four regression tests in |
grantfirl
approved these changes
Dec 16, 2021
…rt_iovr4 * 'main' of https://github.com/NCAR/ccpp-physics: change snow thermal conductivity to verseghy change eta0 to slow snow compaction remove comments and add back the missing if (.not. lmfshal) then block fix the tsurf issue in Noah MP driver update GFS_phys_time_vary.scm.F90/meta to work with latest ccpp/physics and SCM Bug fixes for getting convectice cloud water into Thompson MP without crashes Modification of momentum roughness length over ice in the surface layer scheme Modification of momentum roughness length over ice in the surface layer scheme Add primary codeowners to all files Add more dependency files, and three missing code owners aligned with if block abort model when external surface emissivity file not existing for iemslw=1 Fixed bug in cloud diagnostics, removed over counting. Add codeowners for files that have known Points of Contact cloud cover change associated with xu_randall
climbfuji
reviewed
Dec 27, 2021
There was a problem hiding this comment.
The changes look good. I will wait for the results of the regression testing before I approve. The changes in this PR should not change the results of any of the existing regression tests, since the only tests that were using the iovr==4 option are removed in the associated ufs-weather-model PR.
climbfuji
approved these changes
Dec 28, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
It is a DTC effort to address issue #1 in #748:
1. The EXP cloud overlap option (iovr=4) is currently only allowed to go down a separate path in the code that is labeled in comments as "HWRF" with the initials "mz" (in radlw_main.F90 and radsw_main.F90). This path follows an older coding structure that derives the correlation parameter, alpha, where it is used and that only allows a constant decorrelation length. This appears to me to be close to what was originally coded for HWRF several years ago and went operational in HWRF in 2018. In other words, the existing implementation of iovr=4 is obsolete. HWRF is currently operational with ER (iovr=5), using coding prepared for WRF and not using ccpp-physics.
The associated PRs:
ufs-community/ufs-weather-model#963
NOAA-EMC/fv3atm#445