Feature/ufs nst#589
Conversation
XuLi-NOAA
requested review from
JulieSchramm,
climbfuji,
grantfirl and
llpcarson
as code owners
climbfuji
left a comment
climbfuji
left a comment
There was a problem hiding this comment.
This looks pretty straightforward, only one change needed as far as I can see.
| real (kind=kind_phys), dimension(im), intent(in) :: oro, oro_uf | ||
| integer, intent(in) :: nstf_name1, nstf_name4, nstf_name5 | ||
| real (kind=kind_phys), dimension(im), intent(in) :: xt, xz, & | ||
| real (kind=kind_phys), dimension(im), intent(inout) :: xt, xz, & |
There was a problem hiding this comment.
If the intent is changed here, also need to change in the metadata (unless the metadata already has intent = inout for these 6 variables.
There was a problem hiding this comment.
If the intent is changed here, also need to change in the metadata (unless the metadata already has intent = inout for these 6 variables.
Dom,
Thanks!
Actually, no need to change intent here (I did that for debugging, I think). Will use (in), instead of (inout), for these 6 variables in sfc_nst_post_run.
|
@XuLi-NOAA This PR is behind the master branch and needs to be updated. |
| type = logical | ||
| intent = in | ||
| optional = F | ||
| [tgice] |
There was a problem hiding this comment.
It seems tgice is set as a constant (271.2). It is not the actual freezing point temperature of seawater but the minimum allowed temperature of sea water (~ -0.054*salinity) assuming a maximum salinity in the ocean of about 36 ppt.
There was a problem hiding this comment.
@XuLi-NOAA didn't make up these names, the were defined previously in GFS_typedefs.meta and CCPP other physics schemes.
There was a problem hiding this comment.
@climbfuji I see. Thanks. I do think the naming is not quite correct in any case.
There was a problem hiding this comment.
If so, then we should record an issue in ccpp-physics or fv3atm to change this in a future commit. Thanks for spotting this!
There was a problem hiding this comment.
If so, then we should record an issue in ccpp-physics or fv3atm to change this in a future commit. Thanks for spotting this!
@climbfuji @DeniseWorthen : I just use tgice which is available, as DOM pointed out. Yes, it is a constant as a safeguard. A salinity dependent frozen temperatures are better or is there an even better way? The naming can be updated when ready.
| ! @{ | ||
| subroutine sfc_nst_post_run & | ||
| & ( im, rlapse, tgice, wet, icy, oro, oro_uf, nstf_name1, & | ||
| & ( im, kdt, rlapse, tgice, wet, icy, oro, oro_uf, nstf_name1, & |
There was a problem hiding this comment.
Adding kdt is not necessary here, but it doesn't matter right now, given that all the regression tests have been run at this time.
There was a problem hiding this comment.
Adding
kdtis not necessary here, but it doesn't matter right now, given that all the regression tests have been run at this time.
@climbfuji : True. It was done for debugging originally, and then I think it is OK to keep it.
There was a problem hiding this comment.
We will remove it again in the future, but it is not worth rerunning all the tests because of such a simple change now.
|
@climbfuji : ready to merge |
|
@binli2337 @XuLi-NOAA merged. Please update your submodule pointer for ccpp-physics and revert your `.gitmodules change in your fv3atm PR. The new hash that the ccpp-physics submodule in fv3atm must point to is 990b9d0 (in the NCAR repository, head of master branch). Thanks! |
* Add convective cloud to radiation, Thompson MP only.
4 participants
Three changes and one bug fix in the NSST model in the coupled frame are ready to merge, see #537.
The new baseline results are at: /scratch1/NCEPDEV/stmp4/Xu.Li/FV3_RT/REGRESSION_TEST_INTEL