The new CCCPP suite FV3_GFS_v16b_ugwpv1 and regression tests for GFSv16b-127L/C96 configuration

[ValeryYudin-NOAA]

Description

The new suite represents the upgrade of uGWP (V1), CU-CIRES. It is capable to reproduce the realistic FV3GFS climate, including equatorial dynamics (QBO and SAO, 20-60 km) and annual cycles in the extra-tropics and polar regions in the upper atmosphere (20-80km). It can be executed and tested with the gsldrag sub-grid scale orography physics of gsl/develop for the UFS-R2O tests in 2021.

Solution

The introduction and commit of new suite FV3_GFS_v16b_ugwpv1 will require 3PRs in:

1. ufs-weather-model ; 2) fv3atm; 3) cccp-physics; for performance of the warm-start

For performance of the cold-start it requires update of ufs-weather-model/FMS/sat_vapor_pres/sat_vapor_pres.F90
(tmaxc=160) to make the cold-start from "unbalanced" analyses during diagnostics of RH-values in the upper layers (mesosphere) for atmos_4xdaily.tile1.nc on the first time step.

Therefore the optional 4PD in ufs-weather-model/FMS maybe considered for the top-lid (127L) UFS.

Alternatives

N/A

Related to

Repositories that will be touched: ufs-weather-model , fv3atm, ccpp-physics
The user's feature branches on
https://github.com/ValeryYudin-NOAA/ufs-weather-model branch ugwpv1_gsldrag
/fv3atm branch ugwpv1_gsldrag
/ccpp-physics branch ugwpv1_gsldrag

It requires also the work on the upgrade of the 127-level INPUT and RESTART data inside:
/scratch1/NCEPDEV/nems/emc.nemspara/RT/NEMSfv3gfs/input-data-20201220/FV3_input_data_127
that can be replaced by user's input directory for FV3-127L configuration
/scratch1/NCEPDEV/swpc/Valery.Yudin/noscrub/CCPP_GEOS/FV3_input_data