Skip to content

Conversation

@matthewhoffman
Copy link
Collaborator

This draft PR moves the ability to remove solid and liquid runoff from MPAS-Ocean where it currently occurs to the data runoff component. This has two primary advantages:

  • In G-cases with MALI using TF coupling and DIB active, zeroing rofi in MPAS-Ocean also zeros the ISMF melt flux coming from MALI as rofi. This would also apply to G-cases with MALI applying calving fluxes.
  • Coupler budgets actually reflect the runoff passed between components, as opposed to MPAS-Ocean receiving fluxes, zeroing them out, and then needing to tally how much flux was moved.

This is implemented by adding namelist options to DROF for removing rofl or rofi from either surrounding Antarctica or Greenland. The activation for disabling AIS rofi/rofl follow the same logic as used previously to activate the config_remove_ais_ice_runoff and config_remove_ais_river_runoff options in MPAS-Ocean. The options for Greenland are not currently used. When these options are active, there is new code in DROF that zeros the corresponding field (rofl or rofi) in the r2x attribute vector before it is sent to the coupler. Thus, it is as if the fluxes in those regions never existed.

This commit is an initial demonstration of modifying rofi and rofl based
on lat/lon values.  It is meant to be replaced by a more appropriate
treatment.
This commit adds new DROF namelist options to remove either rofl or rofi
from around either the Antarctic Ice Sheet or the Greenland Ice Sheet.
This is to be used when those runoff fluxes are represented elsewhere -
either by data icebergs or an active GLC component.

Logic is added in DROF's buildnml to activate the AIS options following
the same rules currently for removing runoff currently implemented in MPAS-Ocean.
The Greenland options are not currently used.
Now that this functionality occurs to DROF, remove it from occurring in
MPAS-Ocean.  Note this does not remove the functionality from
MPAS-Ocean, just disable the logic that activated the corresponding
namelist options based on compset.
@matthewhoffman
Copy link
Collaborator Author

Testing

Initial testing using first commit on this branch. As a proof of concept, DROF is hard-coded to zero out any locations not between 0-90 deg longitude. Tested with ./create_test --wait --walltime 0:30:00 SMS_D_Ld5.TL319_oQU240wLI.GMPAS-JRA1p5.chrysalis_gnu

Here is where rofl>0. It matches expectation of only being nonzero between 0-90 deg longitude.

image

@matthewhoffman
Copy link
Collaborator Author

matthewhoffman commented Aug 1, 2025

Testing

Using df9a991. This version has implemented the runoff removal in DROF based in case options and uses the same latitude limits around Antarctica as MPAS-Ocean had previously done. It also disables the runoff removal in MPAS-Ocean, so effects below must be coming from the changes to DROF. Test is run with ./create_test --wait --walltime 0:30:00 SMS_D_Ld5.TL319_oQU240wLI.GMPAS-JRA1p5-DIB-PISMF.chrysalis_gnu. This compset has:

  • rofl removed around Antarctica; replaced by ISMF in MPAS-Ocean
  • rofi removed around Antarctica; replaced by DIB

Images show this is behaving as expected:
image
Note that riverRunoffFlux is unmodified outside of Antarctica. To confirm that iceRunoffFlux is also unmodified outside of Antarctica,
here it is around Greenland:
image

This avoids doublecounting the ice-shelf melt flux calculated from MALI.
If MALI is active and in prognostic mode, use calving flux from MALI and
disable DROF's rofi coming.  Logic is included to differentiate between
ice sheets (including the possibility of a AISGIS MALI mesh).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant