Test mom-ocean/MOM6 PR #1653 #219
Conversation
|
!test repro |
|
❌ The Bitwise Reproducibility Check Failed ❌ When comparing:
Further informationThe experiment can be found on Gadi at The checksums generated by this The checksums compared against are found here https://github.com/ACCESS-NRI/access-om3-configs/tree/3e4442571b3bcd3ac3df759d743130885be49bfa/testing/checksum |
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
dougiesquire
force-pushed
the
test-1deg_jra55do_ryf-mom-pr1653
branch
from
d2d27a5 to
0ca97a9
Compare
|
!test repro |
|
❌ The Bitwise Reproducibility Check Failed ❌ When comparing:
Further informationThe experiment can be found on Gadi at The checksums generated by this The checksums compared against are found here https://github.com/ACCESS-NRI/access-om3-configs/tree/3e4442571b3bcd3ac3df759d743130885be49bfa/testing/checksum |
|
Answer changes in ACCESS-OM3 come from these three commits:
The first two are changes we want. The glc runoff fluxes are part of a large group of new fields coupled when Ping @chrisb13, @anton-seaice, @minghangli-uni, @ezhilsabareesh8 |
|
Thanks @dougiesquire. Totally agree on introducing our access-coupled flag! I haven’t looked deep into the PR related to the glc runoff fluxes, but my understanding is that we’re not using glc in our configurations. So why would this change our answers? |
That's right, in our case the glc runoff flux arrays are all zero. But the inclusion of the zero-valued flux arrays changes answers (by some tiny amount). |
Being able to seperate the liquid and frozen portion of runoff could be useful for future flexibility. We could distribute frozen runoff spatially and liquid runoff at the coast. I think this might be useful for CM3 also. (To use with JRA DROF,. it would require modifications (see https://github.com/COSIMA/access-om3/issues/231#issuecomment-2594595685) as liquid runoff is included currently with river runoff. ) This commit also adds enthalpy from runoff fluxes. It's something that folks have experimented with in OM2 but not been that convinced about the outcomes. I suspect at some point it will be useful. If it just makes a rounding error change to answers its fine to accept the change I think |
|
The glc fluxes are not the only newly added fluxes. There are also a large set for coupling MARBL. I think a meeting would be helpful.
|
… fields if MARBL is in use
|
!test repro |
|
❌ The Bitwise Reproducibility Check Failed ❌ When comparing:
Further informationThe experiment can be found on Gadi at The checksums generated by this The checksums compared against are found here https://github.com/ACCESS-NRI/access-om3-configs/tree/3e4442571b3bcd3ac3df759d743130885be49bfa/testing/checksum |
Do not merge
This is a temporary PR to test the changes to MOM6 included in mom-ocean/MOM6#1653. Companion to ACCESS-NRI/ACCESS-OM3#63