Update of subgrid-scale cloud physics for MYNN/GF#2
Conversation
|
@haiqinli I created a placeholder for the proposed changes (your original PR climbfuji#2 update to the head of gsd/develop and pointing to gsd/develop's new home on the NOAA-GSD GitHub page). This PR must be accompanied by changes in fv3atm (at least the suite definition files and the CCPP prebuild config because of the change of the scheme names from MYNNrad_pre to SGSCloud_RadPre, same for post). Does this PR exist somewhere? |
|
@climbfuji I just modified the SDF and prebuild config in my local copy, and have not created this PR. Do you want me to create a PR for this? |
In this case it is ok if you give me access to the files you have changed, and I will create the FV3 PR. Thanks! |
|
@climbfuji I will send you the path of these files on Hera when Hera is back. Thanks. |
|
@climbfuji The CCPP prebuild config file is /scratch2/BMC/gsd-fv3-dev/Haiqin.Li/gsd-ccpp_heraT/sorc/fv3gfs_ccpp.fd/ccpp/config/ccpp_prebuild_config.py. The SDFs are at /scratch2/BMC/gsd-fv3-dev/Haiqin.Li/gsd-ccpp_heraT/sorc/fv3gfs_ccpp.fd/ccpp/suites. Thanks a lot. |
Thanks. Can you please let me know which of the suite definition files you changed? All of them are different from what it is gsd/develop today, because your code base is far behind. Thanks! |
|
@climbfuji The followings SDFs are changed. Thanks. |
Some of these don't exist in the current gsd/develop code. I updated the following SDFs: Basically, I replaced |
|
@climbfuji Great! Thanks. |
|
@haiqinli I seem to be missing some changes in your fv3atm repository. ccpp_prebuild.py fails with the error message Can you please provide the missing updates to |
|
@climbfuji I am sorry that I forgot to commit this change. Yes, only the update of qci_conv. It is in /scratch2/BMC/gsd-fv3-dev/Haiqin.Li/gsd-ccpp_heraT/sorc/fv3gfs_ccpp.fd/FV3/gfsphysics/GFS_layer/GFS_typedefs.F90 (meta). |
|
Thanks. I will make slight changes (only allocate the array when GF deep convection is used and modify the variable declaration in |
…f instead of hardcoded value 3
…an optional argument for progcld5, because it may not be allocated
|
@haiqinli @joeolson42 @tanyasmirnova please review this PR and post your comment here (I cannot add you as official reviewers). Thanks. |
haiqinli
left a comment
There was a problem hiding this comment.
It looks good to me. Thanks.
…reitas convection scheme is used
d1349cd to
7289051
Compare
|
At last! Regression testing on hera/intel using rt_ccpp_gsd.conf after all changes succeeded (first create, then verify): all tests pass, including the warmstart test. |
|
@tanyasmirnova @haiqinli @joeolson42 can you please confirm and approve that this version is good to merge? Thank you! |
|
Superseded by #12 |
CHSP latlon post: add missing code updates to CCPP
Increase length of message and update declaration in qcmxmn to avoid writing out of bounds
update from NCAR's master
…s_20200619 Update support/HAFS from NCAR dtc/hwrf-physics 2020/06/19
CCPPv5.0.0 scientific doc sync
merge NCAR/ccpp-physics/master
…Aerosols_dom_20210823_2 Hannah gf radiation update revert aerosols dom 20210823 part 2
…210729 Update "constant" branch with latest changes in authoritative repositories
…te_from_main_20211122 New Thompson cloud fraction update from main 2021/11/22
edit module_mp_thompson.F90 to remove optional keyword for rand_pert …
Minor changes to prepare for merge
Coupling Merra2 aerosol climatology and GOCART forecasted aerosol with Thompson microphysics
move setting of flag from run to init phase
…w_fix Fix to canopy arrays definitions: add intent
This PR updates the subgrid-scale cloud calculation for MYNN/GF and renames the modules from
mynnrad_{pre,post}tosgscloud_rad{pre,post}for future use by other physics schemes than MYNN. The PR also adds a new variable required for GSDCHEM coupling, qci_conv to the subgrid-scale cloud calculation and the Grell-Freitas deep convection scheme.Associated PRs:
#2
NOAA-GSL/fv3atm#3
NOAA-GSL/ufs-weather-model#2
For regression testing, see NOAA-GSL/fv3atm#3.