The model fails when d_con is zero and skeb is True because heat_source is unallocated. This change allocates heat_source when skeb is True. PR submitted and approved in GFDL_atmos_cubed_sphere
set SKEB to T and d_con to zero, and the ATM model will fail. Found when running ensembles in hydrostatic mode on hera using intel compilers.
Additional context
No issue previously assigned.
This change is needed to run SFS ensembles, is one-line change, and is currently not tested in the RTs. Adding this change to another PR for a quick inclusion to ufs-weather-model would be ideal for the SFS testing.
PR submitted and approved in GFDL_atmos_cubed_sphere
Description
The model fails when d_con is zero and skeb is True because heat_source is unallocated. This change allocates heat_source when skeb is True. PR submitted and approved in GFDL_atmos_cubed_sphere
https://github.com/NOAA-GFDL/GFDL_atmos_cubed_sphere/pull/351
To Reproduce:
set SKEB to T and d_con to zero, and the ATM model will fail. Found when running ensembles in hydrostatic mode on hera using intel compilers.
Additional context
No issue previously assigned.
This change is needed to run SFS ensembles, is one-line change, and is currently not tested in the RTs. Adding this change to another PR for a quick inclusion to ufs-weather-model would be ideal for the SFS testing.
PR submitted and approved in GFDL_atmos_cubed_sphere
https://github.com/NOAA-GFDL/GFDL_atmos_cubed_sphere/pull/351
Output
change is at line 386 in dy_core.F90
Testing:
Have you tested the code changes? On what platforms? Code has been tested in g-w on hera.
Have you run regression test in ufs-weather-model or ufs-s2s-model with code changes? RT testing in process, no expected changes in baselines.
Dependent PRs: