* Add initialization for some local variables.

* Change undefined value of aextc55 as 0

* Read soill to level 9 for RUC LSM

* Correct tke reading and update upp revision.

* Update upp revision to baa7751

* If the output bundle has no fields do not attempt to get a grid
* Comment out return status check after recover_fields

* Add option to output top parent history file on cubed sphere grid

* Rename module_configure parameter, history_file_on_native_grid

* switch sign on mean momentum fluxes

* send mean momentum fluxes to mediator w/ correct sign for coupling
* fix variable alignment in module_block_data

* change sign of exported sensible heat flux

* add evap fields for atm export

* change sign for exported latent heat flux

* point to Greg's bugfix branch: few small fixes to Thompson MP

…#681)

* Changes to logging and initialization of the CLM Lake Model.

1. Use ice thickness hice(i) to find the level in the lake where ice is
   zero.
2. Do not allow lake temperature to be below freezing point if there is
   no ice.
3. If there is no snow or ice, do not allow surface lake temperature to
   be below freezing point.
   These changes fixed the problem with large errors in the energy budget
   at the beginning of the cold-start run with lakes.
4. Added flag to turn on debug print statements in the CLM lake model.

* merge ccpp-physics NCAR#91 (UFS-SRW v3.0.0 SciDoc updates)

… & FV3_HRRR_gf (NOAA-EMC#692)

* fix fortran coding error in dynamical core
* use clm lake in fv3_hrrr_c3
* initialize arrays after allocation
* ressurect FV3_HRRR_gf suite and give it the clm lake model
* bug fix from Dusan to use the correct type kind when reading lan & lon in quilt server

…A-EMC#595)

* Change the format of domain restart files to netcdf4 and set chunksizes

* Remove nc_format="netcdf4" argument when opening restart files

* Set chunksize of zaxis and time axis to 1

* Update clm_lake and rrfs_sd modules to support chunksizes

* Make quilting restart files identical to fms files

* 2D advection of cellular automata

* update missing value

* adding timing information

* add write_runtimelog option

* update upp and not include dycore updates that change results

* add SPP support to G-F deep convection

* MYNN SFC OpenACC directives

…nd used for data variables. (NOAA-EMC#697)

* Write netcdf axis variables using the same real kind as data variables

…sics, and string length mismatch in dycore plus PR NOAA-EMC#705 and NOAA-EMC#699 (NOAA-EMC#695)

* GFDL_atmos_cubed_sphere: consistent string lengths in array

* stop FV3_HRRR_c3 from crashing with gnu debug

* in dycore, initialize srf_wnd_var2 and tracers_var3 arrays

* Write netcdf axis variables using the same real kind as data variables

* "GF radar reflectivity, dust bug fix, C3 updates, more fluxes output"

* Fix race condition in GFS_phys_time_vary.fv3.F90 error detection

* More bug fixes to GFS_phys_time_vary.fv3.F90:
1. detect empty errmsg from subroutines
2. Initialize err variables in set_soilveg.f, which is called from GFS_phys_time_vary.fv3.F90

* initialize errmsg & errflg in noahmp_tables.f90

* only read h2odata, ozdata and noahmp table when they are needed

* remove all constant 3D variables from clm lake

* calculate clm lake constants only once per i loop

---------

Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov>
Co-authored-by: Haiqin.Li <Haiqin.Li@noaa.gov>

* change mean->inst for fields going to ATM
* switch 'mean' prec fields to be inst

* bug fixed for mraerosol

* update upp revision

* allow cpllnd to use inst lw and sw down

* change print statements to use ufs, not nems

* update active attributes in GFS_typedefs.meta

* add flag_for_land_coupling to active value for both
dlwsfci_cpl and dswsfci_cpl

* update ccpp and revert gitmodules

* Added 'ugwpv1' suite definition files and fixed ccpp UGWP OGWD stability issues in upper atmosphere

* Corrected new 'ugwpv1' SDF's

* Updated ccpp-physics to incorporate PR#115 -- 'hr3_land_upgrades'

* Add zstandard compression

* Enable shuffling with zstandard compression

* Refactor io/module_write_netcdf.F90

* FV3 changes for refactored ozone physics scheme

* Remove change to metadata (Not relevant for PR)

* Split ozone physics into time_vary and run components

* Renamed file. Update SDFs

* Update ccpp-framework hash

* Updates to support NSSL microphysics in CCPP

* Switched atmos and ccpp modules to NSSL versions

* Made setting of otsptflag more logical

* change dimensions of flag_convective_tracer_transport_interstitial to match allocation/use

* Add flags and code for 3-moment rain/graupel/hail in NSSL microphysics scheme

* Update otsptflag for 3moment variables