search

COSIMA / access-om3

ACCESS-OM3 global ocean-sea ice-wave coupled model
13 stars 7 forks source link

Sea Ice BGC Notes for CICE6 #187

Open anton-seaice opened 3 months ago

anton-seaice commented 3 months ago

Nicole Jeffery reports that the "Skeletal layer BGC is not working" in latest / current Icepack (and therefore CICE). It may just be deprecated because we don't think anyone is using it, as its been replaced by the multi-layer BGC.

Does this sound ok? Do we think we want this to run future comparisons to OM2 results ? @pwongpan @aekiss

aekiss commented 3 months ago

I hadn't envisaged using the old scheme in CICE6 (I assumed we'd go straight to the multi-layer version), but maybe @pwongpan had different plans?

pwongpan commented 3 months ago

@aekiss @anton-seaice Thanks again, Anton, for the updates. I agree with Andrew we will go straight to the multi-layer BGC to improve, in particular, summer sea ice algal phenology.

BTW, I would like to try zbgc option in ACCESS-OM2 BGC which is also available also on CICE5 we are using (see a screenshot below).

Screenshot 2024-07-12 at 9 31 42 AM

@aekiss did you experiment turn this on before?

Then we could compare the two multi-layer schemes between ACCESS-OM2 and ACCESS-OM3. This could be a great comparative opportunity. What do you think? Thanks again Team.

aekiss commented 3 months ago

The only ice BGC experiment I've been involved with is 01deg_jra55v140_iaf_cycle4, which used these zbgc parameters: https://github.com/COSIMA/01deg_jra55_iaf/blob/01deg_jra55v140_iaf_cycle4/ice/cice_in.nml#L135-L153

tr_brine = .false. so it looks like zbc wasn't activated.

ofa001 commented 3 months ago

There is the complication for ESM3 when coupled to the UM that we will not be able to use Mushy layer ice but still have to use the bl99 option for ACCESS-CM3 coupling. Though @pwongpan did discuss last year that in the longterm there may be a way to couple mushy in its not the way we are doing it at the moment as its much more complicated. This may be an issue to think about in these discussions. But I am sure the first step is to compare how ice BGC working in ACCESS-Om2 v ACCESS-OM3

aekiss commented 3 months ago

There are also technical issues to be overcome before ACCESS-OM3 can do ocean-ice BGC coupling.

anton-seaice commented 3 months ago

FYI @pwongpan @aekiss - There are some tweaks going into Icepack (and therefore CICE) re congelation ice formulation when using mushy thermodynamics. It accounts for the liquid fraction when forming new ice through congelation, and therefore reduces the basal heat flux (i.e. less heat is removed from the ocean) which should give reduced (i.e more realistic?) frazil formation.

Details are in the Plante et al 2024 paper and https://github.com/CICE-Consortium/Icepack/pull/494

pwongpan commented 3 months ago

@anton-seaice @aekiss Thanks again for the heads up. Do we already have the option to tweak the turbulent ocean-sea ice nutrient exchanges according to Duarte et al. (2022) too? I think these two implementations could go along together.

aekiss commented 3 months ago

Thanks @anton-seaice, sounds interesting.

@pwongpan do you know whether Duarte's code changes have been merged into the main codebase? I searched for Bottom_turb_mix in CICE and Icepack but didn't find it.

pwongpan commented 3 months ago

@aekiss It seems to be a fork hosted on Zenodo, not in the main codebase: https://zenodo.org/records/5795034 with "This code is in a fork derived from the CICE Consortium repository (https://github.com/CICE-Consortium, last access: 26 January 2022).".