Running historical and ssp585 CM2 simulations with meltwater forcing

Earth System
,
1

Hi,

I aim to run some historical and ssp585 simulations with additional Antarctic meltwater input using ACCESS-CM2. The u-db130/meltwater suite has the required code to apply the meltwater perturbation.

I was wondering if someone could help me with the following:

  • What suite to choose for the historical/ssp585 configurations?
  • How to add the meltwater code to these suites?
  • How to control time-varying meltwater forcing? I have so far only experience in running simulations under constant forcing. I’m assuming the time-varying meltwater can be controlled in a similar way to how historical/ssp585 conditions evolve.

Many thanks,
Wilma

2

@wghuneke , I’m on triage today so I had a quick look to see if we had enough information.

Have you read the README from the u-db130/meltwater suite:

edit suite.rc to get new ice code for the extra waterflux to be passed into ocean as iceberg (melting latent heat into account) or "runoff" (actually anywhere as assigned, just waterflux itself)

edit app/cice/rose-app.conf to add control for applying the extrafw
[namelist:coupling]
......
do_extrafw=.true.
extrafw_as_iceberg=.false.
extrafw_factor=1.0

edit app/coupled/rose-app.conf to set the extrafw datafile to be read in by CICE
[file:$ICE_RUNDIR/INPUT/extrafw.nc]
source=/g/data/p66/dhb599/ACCESS-CM2/input/cice/Extrafw/AA_extrafw.nc

then, run the suite:
rose suite-run --new

This seems to list the controls and the input file, I’m not sure if you are looking for more fine grain controls or not. I would assume the interannual variability is controlled by the input file, but that would need to be checked or confirmed.

Looking into the suite.rc file itself, based on comments, it seems the code is added during the make_cice step:

    [[make_cice]]
        inherit = BUILD, NCI
        script = """
if [ -d $ICE_ROOT ] ; then
  /bin/rm -fr $ICE_ROOT
fi
svn co file:///g/data/access/access-svn/cice/branches/access/cice_gsi8.1@404 $ICE_ROOT
#B: need to replace a few (code) files (the new ones have been modified for the extra waterflux experiments):
cp /g/data/p66/dhb599/ACCESS-CM2/submodels/cice5_access2_EXTARFW/drivers/access/* $CYLC_SUITE_SHARE_DIR/cice/drivers/access/
"""

I can’t say for sure if it’s the only difference (although it would make sense that it would be).

For the suite ID for the historical and ssp585 configurations, I have no idea. I will ask or someone might see this and pipe in.

3

@dhb599 It seems you may have created the original suite with the extraflow that @wghuneke wants to use. If you have anymore information, please feel free to jump in.

4

Thanks @clairecarouge for your input and apologies for the delayed reply. I’m familiar with the code pieces you shared, I can add them to the respective suites.

Does anyone know which CM2 historical and ssp585 suites to use?

And does anyone have experience in running those suites, e.g. how to best handle time-evolving input fields?

Thanks,
Wilma