Conversation
…y, additional code updates and bugs
hannahcbarnes
requested review from
SamuelTrahanNOAA,
christinaholtNOAA,
haiqinli,
joeolson42,
mdtoyNOAA,
middlecoff,
tanyasmirnova and
zhanglikate
as code owners
|
@haiqinli Can you look double check these modifications? |
|
@hannahcbarnes The variable of aod_da should be added into cu_gf_driver.meta. The others look good. Thanks. |
|
@haiqinli thanks. aod_da should now be a part of cu_gf_driver.meta |
- Minor fixes to pass regression tests
physics/sgscloud_radpre.F90
Outdated
| real(kind=kind_phys), dimension(:,:), intent(inout) :: qci_conv | ||
| real(kind=kind_phys), dimension(:,:), intent(in) :: ud_mf | ||
| real(kind=kind_phys), dimension(:,:), intent(in) ::ud_mf | ||
| real(kind=kind_phys), dimension(:,:), intent(in) ::ud_mf_timeave, qci_conv_timeave |
There was a problem hiding this comment.
minor detail but can we preserve the space between :: and the first variable?
| endif | ||
| sigq = 1.5E-3 * ud_mf(i,k)/dt * f | ||
| sigq = ud_mf_timeave(i,k)/dt * f | ||
| !sigq = 1.5E-3 * ud_mf(i,k)/dt * f |
There was a problem hiding this comment.
This is a bit shocking. Are you sure sigq will be ~O(1e-4) without this constant? If sigq is too big, you might always get cloud fractions of around 50%,
There was a problem hiding this comment.
I can check. I have output arrays of just GF cloud fraction. Where there are clouds, cloud fraction values tend to be between 30-70%.
joeolson42
left a comment
joeolson42
left a comment
There was a problem hiding this comment.
I'm OK with the changes so long as the GF developers are sure they are desired. I left a couple questions just to make sure.
5 participants
Enables GF to use AOD analysis instead of AOD from MERRA2 climatology
Fixes GF - radiation coupling
Fixes dry / moist mixing ratios in GF
Additional tuning to GF
Reverts previous change to sgscloud_radpre
The regression test logs are in /scratch1/BMC/gsd-fv3-dev/hcbarnes/CommitTest/sorc/ufs_model.fd/tests/