[PULL REQUEST] Forward and reverse models compile and run#67
[PULL REQUEST] Forward and reverse models compile and run#67lizziel merged 5 commits intogeoschem:devfrom
Conversation
yantosca
changed the title
msulprizio
added
category: Feature Request
|
Hi @TerribleNews, thanks for these updates. I am bringing these in for the next release of HEMCO. I have similar questions as for the GEOS-Chem companion PR.
It's been on the to do list for a while to improve the error handling and prints in HEMCO, especially when using MPI. If some or all of your prints were for temporary debugging we might still want to bring them in, but maybe with tweaking if they are MAPL-dependent. |
Okay, I'll look into this and get back to you. What is the plan for debug prints?
Oh, that's a good question. I've never build GEOS-Chem classic from the GCHP repo. I will do a quick visual check, but it might take me a while to get to testing if I can build (and run?) classic. |
|
I think I've covered most of these now. For now I just eliminated the debug printing altogether. It shouldn't be hard to add back in once there's more infrastructure for handling MAPL/non-MAPL cases. |
|
Thanks @TerribleNews! I'll bring this in this week and let you know if I have any questions. |
|
|
||
| END SUBROUTINE GetHcoDiagn | ||
|
|
||
| #ifdef FALSE ! the adjoint-relevant code in here has been moved to |
There was a problem hiding this comment.
Did you intend to leave this as FALSE?
There was a problem hiding this comment.
Sorry, I missed that when I cleaned up the other ifdef FALSEs. This whole subroutine was moved to hco_calc_mod but I left it there as insurance in case that didn't work and then forgot to delete it after I verified the new code. I have commited a change removing both these blocks.
| PUBLIC :: SetHcoTime | ||
| PUBLIC :: GetHcoVal | ||
| PUBLIC :: GetHcoDiagn | ||
| #ifdef FALSE |
There was a problem hiding this comment.
Same question as above. Should this be ADJOINT?
src/Core/hcoio_write_esmf_mod.F90
Outdated
| IF (MAPL_am_I_Root()) THEN | ||
| IF ( RC == HCO_SUCCESS .and. FLAG == HCO_SUCCESS ) WRITE(*,*) "Got it! Name: ", TRIM(ThisDiagn%cName) | ||
| IF ( RC /= HCO_SUCCESS) WRITE(*,*) "Fail! RC: ", RC, " Flag: ", FLAG | ||
| ENDIF |
There was a problem hiding this comment.
In my GCHP test this is printing lines every timestep.
There was a problem hiding this comment.
Those wound up being pretty key debug messages. I don't know if we'll need them again, but probably. Can I comment them out but keep them?
There was a problem hiding this comment.
Yes. We want to overhaul the HEMCO debug prints and if this stays in but commented out it will be included in that update.
4 participants
Companion pull request to geoschem/GCHP#71
I needed to add HEMCO code to calculate emissions sensitivities, apply emissions scaling factors, and output adjoint diagnostics.