Updated meta data tables for gfdl cloud microphysics#80
Updated meta data tables for gfdl cloud microphysics#80kellylittleblackdog wants to merge 5 commits into
Conversation
…t routines for init and run. Renamed driver run. Added metadata tables.
…here needed to gain successful compile.
| !! | errflg | error_flag | error flag for error handling in CCPP | flag | 0 | integer | | out | F | | ||
| !! |-------------|---------------|---------------------------------------------|---------|------|-----------|--------|--------|----------| | ||
| !! | ||
| subroutine gfdl_cloud_microphys_pre_init(do_shoc, errmsg, errflg) |
There was a problem hiding this comment.
Currently CCPP does not have the capability to run the scheme _init and _finalize routines as part of the and steps. We need to add this so that this routine gfdl_cloud_microphys_pre_init will be executed.
|
|
||
| if(do_shoc) then | ||
| write(errmsg,'(*(a))') & | ||
| & 'SHOC is not currently compatible with GFDL MP -- shutting down' |
There was a problem hiding this comment.
I suggest to remove the " -- shutting down" part of the error message as it is up to the host model to decide what to do if CCPP returns with ierr /= 0.
|
|
||
| !!===================================== | ||
|
|
||
| module gfdl_cloud_microphys_mod |
There was a problem hiding this comment.
This module must be named "gfdl_cloud_microphys" to match the names of its _run, _init and _finalize routines.
| ! --- interface variables (output) | ||
| real(kind=kind_phys), intent( out) :: seconds | ||
| real(kind=kind_phys), intent( out), dimension(:,:) :: land, area, rain0, snow0, ice0, graupel0 | ||
| real(kind=kind_phys), intent( out), allocatable, dimension(:,:,:) :: delp, dz, uin, vin, pt, qv1, ql1, qr1, qg1, qa1, qn1, qi1, & |
There was a problem hiding this comment.
Several comments here: (1) where are these arrays deallocated? (2) by definition we cannot allocate arrays in a scheme that get used in other schemes (and pre (gfdl_cloud_microphys_pre) and main (gfdl_cloud_microphys) are different schemes). These variables have to become interstitial variables that get allocated in GFS_typedefs when GFDL physics are run (check with namelist/SDF).
When later the automatic flipping of arrays is implemented and automated inside the scheme caps, we can get rid of the extra variables that deal with the flipping.
@ligiabernardet @llpcarson @grantfirl What is your opinion on this?
|
|
||
| real :: missing_value = - 1.e10 | ||
|
|
||
| logical :: module_is_initialized = .false. |
There was a problem hiding this comment.
As just discussed, we can keep these as long as they are not shared with any other model and they do not depend on the spatial dimension im/ix/levs.
|
It appears that an argument table for gfdl_cloud_microphys_finalize is needed. Also, it appears that the names for the three argument tables in the gfdl_cloud_microphys_post module are incorrectly called "pre" instead of "post". |
|
@kellylittleblackdog how is the current status of the GFDL MP? Can you update this PR and those for the other repositories with what you have so that I can start integrating it, please? Thanks, Dom |
|
@kellylittleblackdog this PR is no longer needed, or? |
|
Dom,
This should be the valid push request.
…On Fri, Jun 29, 2018 at 9:04 PM, Dom Heinzeller ***@***.***> wrote:
@kellylittleblackdog <https://github.com/kellylittleblackdog> this PR is
no longer needed, or?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#80 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ALbmrllpFnZ6nn5E7D39vS4ndyH42283ks5uBurMgaJpZM4UD3JI>
.
--
Donald Stark
NCAR RAL JNT/DTC
Address: FL3-1042
Phone: 303.497.2472
Email: stark@ucar.edu
Web: www.rap.ucar.edu/~stark
|
|
Replaced by 108 |
…merged_into_gsl_develop Update thompson mp 20210213 (NCAR#567) for gsl/develop
3 participants
Please review new variables requested by meta data table.