Raising visibility of cavity plans in OM3

Cryosphere
,
1

Hi everyone,

After the Cryo WG meeting today I had a few questions rattling around in my mind - things like:

  • Will ACCESS-OM3 have sub shelf cavities?
  • If so, will it have thermodynamics (e.g. melt)?
  • Could it even have dynamics?

We (Hobart Cryo people) talked a bit about these things after the meeting together. I had naively assumed OM3 wouldn’t have a cavity, so I was mentally planning for fast ice in the model configuration using the ice sheet model, which would have been super expensive.

But we grabbed @PSpence as he walked past our meeting room. He said…

  • OM3 will have sub ice shelf cavities (this is part of MOM6 code)
  • It will have thermodynamics (e.g., sub shelf melt) and even dynamics (e.g. grounding line retreat) - but without the fully coupled ice sheet model expense
  • Cavities will be implemented in both the pan-Ant configuration (1/20th to 1/40th deg) as well as the global model (1/10th I guess?). But not in the regional configuration of OM3 (which has to be north of the ice edge anyway).

Anyway I guess the purpose of this post is to draw attention to this capability within the cryo group.

(@anton @Chen_Zhao @pat_wongpan were in the room and @dpath2o you might be interested too. Paola Papapetros was there too but I can’t find her @)

@PSpence please correct anything I’ve messed up!

Cheers, Alex.

2 Likes
2

Hi @adfraser Yes I knew most of those answers, I guess for climate model runs we see 1/4 degree as the affordable option for ocean resolution for CMIP7 time frame For coupling to the ice sheet I guess the ice cavities will need to be at 1/10 to do a reasonable job. From a fast perspective and @dpath2o project having fast ice and icebergs as well would be ideal to be linked to this work. We can talk offline about how this fits in with your plans as well.

2 Likes
3

Yes that’s the plan @adfraser.

I’d estimate we’re perhaps (at least?) a year away from having cavities in the panan config. MOM6 cavities currently only work with isopycnal vertical coords, but GFDL are working on improving the spurious velocities in ALE coords.

I don’t know of anyone in COSIMA who is planning to use moving grounding lines, so that might take some effort by whoever first wants to use dynamics to configure. MOM6 apparently has the capability for this, but we haven’t tested it.

4

Thanks @adele157 - good info!

@PSpence mentioned the names of the people working on panant but I forgot - could you please let me know again?