Implementing Glacier->Ocean coupling on multiple height levels

[Katetc]

Several people have worked on implementing the transfer of heat and salinity information from the ocean to the ice sheets in CESM in various ways. In this discussion, let's begin by pointing to the various code repositories with work that has been done and describing the state of that code (issues, age, work completed and work to be done, etc). Any documentation on this work, including relevant papers and presentations can be linked and discussed here as well.

Our eventual goal is to support Antarctic ice sheet coupling on multiple depth levels with various ocean components in CESM3 and NorESM.

Those initially involved in this discussion are:
@Katetc @mvertens @billsacks @gunterl @whlipscomb @mpetrini-norce @hgoelzer

Others are very welcome to join!

[whlipscomb]

Here are some google docs I found from 2019–2020, when several of us were first working on this coupling:

(1) Sarah's notes on ice sheet–ocean coupling for LGM runs.

(2) Gunter's notes on using ocean coupled fields to compute thermal forcing in CISM.

(3) My notes on modifying CISM to ingest ocean T and S from the coupler.

I think all three docs should now be accessible to anyone with the link.

[mpetrini-norce]

Here is another Google Doc I made at the very beginning of our POP-CISM coupling work (it includes several comments by Bill L.); it contains a detailed explanation of the separate vertical levels approach (seven levels, and seven mapping files), some thoughts on remapping (nearest-neighbour vs bilinear+extrapolation) and some paths/routines. It's an older document than 1)-2)-3) and probably outdated, but maybe still useful?

[mpetrini-norce]

Here is the path on Derecho of the latest code version me and Sarah Bradley used to run tests (BG and T-compset) with POP-CISM coupling: /glade/u/home/mpetrini/mycesm2sandbox/cesm-2.1.1_cism_pop

[mvertens]

@mpetrini-norce - do you have any POP output files from these runs - that could be a starting point for a new data ocean mode?

[mpetrini-norce]

I don't have outputs unfortunately - what I have is the scripts to:
(1) create first a fully coupled case with ocean forcing (and to save coupler history files):
/glade/u/home/mpetrini/cesm2cism2_oceancoup/create_PI-BG-pop.csh
(2) create a T-compset using coupler history files saved in (1) (both ocean and SMB)
/glade/u/home/mpetrini/cesm2cism2_oceancoup/create_PI-TG-pop.csh

Hope this can be somehow helpful.

[mvertens]

One option is to use the forced ocean/sea ice CESM2 simulation from OMIP2.
The CESM OMIP2 output can be found at:
/glade/collections/cmip/CMIP6/OMIP/NCAR/CESM2/omip2/r1i1p1f1
The MISOMIP had an idealized domain and used MOM6 instead of POP.