Develop#13
Conversation
… limited area model configurations
…. This version contains GF retuning and a few more changes for limited area runs
egkhamis
left a comment
egkhamis
left a comment
There was a problem hiding this comment.
We found some issues that was commented and need to be fixed:
- the missing variables lh_spr and hfx_spr must be included in the diag_physics pool
- the variable rqvcuten is missing in the Registry.xml
- The variable rqvcuten is missing in the mpas_init_atm_cases.F code
| call mpas_pool_get_array(diag_physics,'znt' ,znt ) | ||
| call mpas_pool_get_array(diag_physics,'zol' ,zol ) | ||
|
|
||
| call mpas_pool_get_array(diag_physics,'lh_spr' ,lh_spr ) |
There was a problem hiding this comment.
the missing variables lh_spr and hfx_spr must be included in the diag_physics pool (call mpas_pool_get_array(...))
There was a problem hiding this comment.
The variable rqvcuten is missing in the Registry:
<var_struct name="tend_physics" time_levs="1">
<var name="rqvcuten" type="real" dimensions="nVertLevels nCells Time" units="kg kg^{-1} s^{-1}"
description="tendency of water vapor mixing ratio due to cumulus convection"
packages="met_stage_out"/>
</var_struct>There was a problem hiding this comment.
This variable does not need to be in init_atmosphere.
There was a problem hiding this comment.
It is needed to create code in compilation time to activate the tend_physics pool, the same will happen with the model. Also, there are problems with init_atmosphere compiled in debug mode, it will not run without this variable (or any other in tend_physics).
There was a problem hiding this comment.
The variable rqvcuten is missing in the code:
real (kind=RKIND), dimension(:,:), pointer :: rqvcuten
call mpas_pool_get_array(tend_physics, 'rqvcuten', rqvcuten)There was a problem hiding this comment.
This variable does not need to be in init_atmosphere.
There was a problem hiding this comment.
It is needed to create code in compilation time to activate the tend_physics pool, the same will happen with the model. Also, there are problems with init_atmosphere compiled in debug mode, it will not run without this variable (or any other in tend_physics).
There was a problem hiding this comment.
Ok. Then, we need to report this potential 'bug' to NCAR people.
There was a problem hiding this comment.
the missing variables lh_spr and hfx_spr must be included in the diag_physics pool:
call mpas_pool_get_array(diag_physics,'lh_spr' ,lh_spr )
call mpas_pool_get_array(diag_physics,'hfx_spr',hfx_spr)
2 participants
This should be the base of the first pre-operational version of MONAN. This version contains GF retuning and a few more changes for limited area runs. It defaults the suite 'convection_permitting_monan' for the pre-operational implementation by the DIPTC. It also contains changes in the 'diagnostic' directory.
To do by the GCC-DMINT: The messages "srf: initial soil moisture reduced by 30%" written in the log file by each mpi task should be written only by the "mpi master,"