Add support for running MARBL with ladjust_bury_coeff = .true. #323
Conversation
Also add support for the three additional surface fluxes that will be requested if ladjust_bury_coeff = .true.
Update control structure to provide memory for glo_avg_fields returned from MARBL when ladjust_bury_coeff = .true.
Note that scalar diagnostics requires diag_ctrl to be intent(inout) so this commit changes MOM_tracer_flow_control.F90 as well.
For fields where MARBL wants running means, we compute the global mean of the 2D field provided by MARBL
Add a few components we use with running means
|
I think I'm going to start a new PR for this work, and will close this PR once the new branch is underway. That might not happen for a little while, though, since I'm going to turn my focus to the other outstanding science issue with MARBL (updating the iron forcing). The plan is to take the commit where we define the new forcing fields (1be59df) and the commit where we add the new diagnostic output (I think that's all of 2b38b5f) and then figure out exactly what MARBL needs MOM6 to do to compute those new diagnostics correctly. The implementation in POP is extremely generic, and includes support for global sums / means and also running means of quantities that MARBL doesn't actually require... so working through the calling tree and only adding fields that we really need to the control structure should make it easier to move forward. |
We need to tune the burial coefficients by running MARBL with
ladjust_bury_coeff = .true.. This setting requires some features that were not implemented in the initial MARBL driver: there's a little more data passed back and forth between MARBL and MOM6 to make sure these parameters are computed correctly.I'll take this out of draft mode and add reviewers when it is ready.