CICE-SCHISM coupling (UFS) updates #215
Conversation
There was a problem hiding this comment.
@SmithJos13 : I don't think u changed GOTM, so can u plz remove this from PR?
src/Hydro/schism_init.F90
Outdated
| allocate(sink0(npa),fraction0(npa),kk10(npa),kk20(npa)) | ||
| #endif | ||
|
|
||
| !#ifdef USE_CICE |
There was a problem hiding this comment.
@SmithJos13: it's good to keep your changes (which are related to CICE6 via UFS) isolated via CPP. I propose UFS_CICE.
src/Hydro/schism_step.F90
Outdated
| !$OMP end parallel do | ||
| #endif /*not USE_ATMOS*/ | ||
|
|
||
| !#ifdef USE_CICE |
src/Core/schism_glbl.F90
Outdated
|
|
||
| !(2,npa). ocean-ice stress (junk if no ice) [m^2/s/s] | ||
| real(rkind),save,allocatable :: tau_oi(:,:) | ||
| !(npa). ocean-ice stress (junk if no ice) [m^2/s/s] |
There was a problem hiding this comment.
Aren't tau_oi[xy] just components of tau_oi?
src/Core/schism_glbl.F90
Outdated
| !>--------------------------------------------------- | ||
| real(rkind), allocatable, save, target :: uvice(:) | ||
| real(rkind), allocatable, save, target :: vvice(:) | ||
| real(rkind), allocatable, save, target :: taux(:) |
There was a problem hiding this comment.
tau[xy] are too close to other existing names. Could we change them to something more distinct like tau[xy]_ice
src/Core/schism_glbl.F90
Outdated
| real(rkind), allocatable, save, target :: taux(:) | ||
| real(rkind), allocatable, save, target :: tauy(:) | ||
| real(rkind), allocatable, save, target :: vsno(:) | ||
| real(rkind), allocatable, save, target :: vice(:) |
There was a problem hiding this comment.
This is also a common name; can we rename it?
|
Thx @SmithJos13. We are almost there. Can u address my question on tau_oi[xy]? |
|
@josephzhang8 Do you want me to remove tau_oi[xy] and stick with tau_oi? |
|
Yes, removing [xy] would be good to conserve memory. Thx |
|
I'll wait for review from @uturuncoglu |
|
Sounds good I'll update and make the consistent changes on schism esmf. we should review both before we Marge in case other vars need to change I'm about to submit the emsf pr |
|
Do I need to use a different exchange routine or is call exchange_p2d(tau_oi) okay? |
SmithJos13
force-pushed
the
feature/cice_updates
branch
from
ae2ea60 to
98de8db
Compare
|
I just removed all the auxiliary commits from the PR |
I see why u declared [xy] now. You can try to allocate those [xy] as pointers and then use them for exchange. This way it won't waste any memory. |
|
Can you please show me where the exchange is done? I can fix it for you using swild* if they are inside SCHISM codes. |
2 participants
Changes to the initial implementation of CICE coupling for current working configuration in UFS-coastal. Bug fixes. Please tell me if you want me to add in CPP_FLAG (USE_CICE) protection commented out currently. Would have implications at the app level in UFS-coastal. @uturuncoglu tagged here as requested.