You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Following discussion in #2861 summarizing questions/comments here:
While the "usual" coupling through homme-sourced ApplyCAM... functions is still active in DPxx, there is also additional IOP-forcings coupling.
The IOP coupling now does not match F version of DP (Exner pressure is computed from phydro in the xx version) and it does not match the "usual" coupling even more (theta and geopotential tendencies are computed differently).
If there is an intent to use IOP studies as a prediction for how the full model will behave, the mismatch in coupling may affect reliability of predictions (the model is sensitive tot he coupling mechanisms).
If DP F and DPxx are supposed to be the same wrt climate, then why does xx version need different Exner definition? It may be unrealistic to have bfb for F and xx versions of DP compsets, but would it be reasonable to have bfb tests for IOP forcings (and whatever else that xx infrastructure introduced)?
(another spin on the item above) If there was more verification done for the F code, then having a different IOP coupling mechanism in the xx version means the models (F and xx versions) may not be close to each other.
What happened to the pressure adjustment in case of IOP forcing? I haven't seen the corresponding code in the xx version (i did not check F version for it). Without it the model introduces a mass leak (dry mass).
The text was updated successfully, but these errors were encountered:
Following discussion in #2861 summarizing questions/comments here:
While the "usual" coupling through homme-sourced ApplyCAM... functions is still active in DPxx, there is also additional IOP-forcings coupling.
The IOP coupling now does not match F version of DP (Exner pressure is computed from phydro in the xx version) and it does not match the "usual" coupling even more (theta and geopotential tendencies are computed differently).
If there is an intent to use IOP studies as a prediction for how the full model will behave, the mismatch in coupling may affect reliability of predictions (the model is sensitive tot he coupling mechanisms).
If DP F and DPxx are supposed to be the same wrt climate, then why does xx version need different Exner definition? It may be unrealistic to have bfb for F and xx versions of DP compsets, but would it be reasonable to have bfb tests for IOP forcings (and whatever else that xx infrastructure introduced)?
(another spin on the item above) If there was more verification done for the F code, then having a different IOP coupling mechanism in the xx version means the models (F and xx versions) may not be close to each other.
What happened to the pressure adjustment in case of IOP forcing? I haven't seen the corresponding code in the xx version (i did not check F version for it). Without it the model introduces a mass leak (dry mass).
The text was updated successfully, but these errors were encountered: