Bromoform coupling to atmosphere

[jmaerz]

Hi @TomasTorsvik and @JorgSchwinger , it seems to me that bromoform earlier was coupled to the atmosphere - exchanging both the fluxes fields and atmospheric bromoform concentration. The latter currently seems not to be case anymore (as of #658). Since I barely have looked into bromoform (and being unaware of the currently expected behavior), I open this issue as a reminder to cross-check. Please feel free to comment or close it again. I likely won't work on it.

There are a number of closed PR attached to exchange of bromoform (not sure, if they help, though):
#360
#264
#259

[mvertens]

@jmaerz - I'll add a few details here about the current state of bromoform.

A long time ago I put the hooks in CMEPS for BLOM bromoform to be coupled to CAM:

in cmeps/mediator/esmFldsExchange_cesm.F90 Faoo_fbrf_ocn is advertised by the mediator and forms a connection to BLOM if BLOM also advertises the field

in cmeps/mediator/med_phases_prep_atm_mod.F90 we have the following module variables

 character(len=13) :: fldnames_from_ocn(5) = (/'Faoo_fbrf_ocn','Faoo_fdms_ocn','Faoo_fco2_ocn',&
                                               'Faoo_fn2o_ocn','Faoo_fnh3_ocn'/)

and these fields are mapped to the atm grid and then scaled by (icefrac+ocnfrac) on the atm grid as long as CAM advertises that it wants bromoform

However, in the CAM nuopc cap - currently the following is hard-wired:

logical, public :: brf_from_ocn = .false. ! brf is obtained from ocean as atm import data
So even if BLOM sends bromoform to the mediator - it will never be sent to CAM.

Also - checking out Noresm2.3 and Noresm2.1 - I do not see bromoform in the CAM nuopc cap - or anywhere in the CAM code.
Was the bromoform coupling done on some other branch?

More importantly - what are the current requirements for bromoform coupling to CAM and do we want to bring the required functionality in CAM to accept bromoform from BLOM?

[jmaerz]

Just to clarify: if I am not mistaken, bromoform was previously coupled via MCT. But better @JorgSchwinger can comment on expected behavior.

[mvertens]

@jmaerz - yes - bromoform in the past was only coupled via MCT. I don't know what version of the NorESM code implemented this coupling in CAM.

[JorgSchwinger]

The Bromoform code was developed during KeyClim, and there was a model version that also coupled the fluxes to the atmosphere (but without a more detailed chemistry, there is not much reason to actually do this).

So at the moment, I think it is fine as is, and I do not think it is a priority to enable the coupling to the atmosphere for NorESM3 @tjiputra @DirkOlivie ?

[DirkOlivie]

Hi, yes this was developed in and used for KeyCLIM. I don't think this is an urgent topic, but at some point we might try to get it into NorESM3. For the atmosphere it should be seen as an additional chemical scheme.

[mvertens]

@DirkOlivie - thanks for the confirmation. Could you create an issue in NorESMHub/CAM as an enhancement outlining what needs to be done to CAM and specifying what KeyCLIM code base had this before we forget? The hooks are already there in CMEPS and BLOM. So CAM is the missing piece.