[Ncep.list.nems.announce] NEMS commit

Gerhard Theurich theurich at sourcespring.net
Tue May 24 16:58:50 UTC 2016


Just to point out that we will need a procedure in place to test coupled 
runs, and update their regression baselines as well when GSM results 
change.

One approach would be that once an email like the one below from Jun 
comes across, the UGCS-Seasonal code manager (and similar for other 
coupled systems) would kick of test runs of "their" coupled system with 
the new GSM revision. Since results have changed, a qualitative analysis 
will need to be initiated. Once that has completed, the trunk of 
UGCS-Seasonal (and similar) can start pointing to the new GSM revision.

Other ideas? Who will do this now for UGCS-Seasonal after this GSM trunk 
commit?

-Gerhard

On 05/24/2016 07:51 AM, Jun Wang - NOAA Affiliate wrote:
> Committed revision: 76291
>
> Ratko,
>
> Would you please copy over the gsm regression results on theia and
> wcoss? Thanks.
>
> Jun
>
> On Mon, May 23, 2016 at 11:42 AM, Jun Wang - NOAA Affiliate
> <jun.wang at noaa.gov <mailto:jun.wang at noaa.gov>> wrote:
>
>     Dear all,
>
>     Some post merge gsm changes are ready to commit to trunk. The
>     changes include:
>
>     1)zhour is zero in nemsio flx files, it needs to be updated in the
>     forecast integration
>
>     2)gsm run script exglobal_fcst_nems.sh needs to add nemsio.configure
>     and atmos.configure to support standalone gsm run (e.g. parallel run)
>
>     3)ATM clock needs to reset stopTime for restart case.
>     src/atmos/module_ATM_GRID_COMP.F90 is updated with Gerhard's fix.
>
>     4)regression test gfs_slg_nemsio_read is added by the big merge,
>     which was replaced by gfs_slg for a while in previous trunk.
>     gfs_slg_nemsio_read is removed
>
>     5)Wam is still using last forecast PE to write out nemsio file, the
>     index for the model level file is out of boundary and is fixed.
>
>     6)In previous fix for diagnostic variable PBL clouds, the si is
>     computed from info in nemsio sigma files, this fix does not apply to
>     the sigma file input. So the piece of code to compute si is added
>     back for sigio/sfcfio input.
>
>     7)TRAP related changes:
>     a) the subroutine dyn/getpwatptot can't be called in reset step as
>     the input fields are different from normal step.
>     b) Mf convection scheme has uninitialized variables.
>     With the two fixes, NEMS gsm can run with TRAP turned on.
>
>     Due to zhour change and the fix of uninitialized array in
>     mfshalcnv.f, new gsm baseline on theia and wcoss are created.
>
>     Regression test passed on both wcoss and theia, ready to commit to
>     trunk.
>
>     The NEMS branch can be found at:
>
>     https://svnemc.ncep.noaa.gov/projects/nems/branches/jun/nems_gsmtrapon
>
>     The gsm branch is at:
>
>     https://svnemc.ncep.noaa.gov/projects/gsm/branches/jun/gsmtrapon
>
>     Please let me know if you have any questions.
>
>     Thanks.
>
>
>     Jun
>
>
>
>
> _______________________________________________
> Ncep.list.nems.announce mailing list
> Ncep.list.nems.announce at lstsrv.ncep.noaa.gov
> https://www.lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.nems.announce
>


More information about the Ncep.list.nems.announce mailing list