Icebergs in ACCESS-OM3?

[adele-morrison]

Not a priority, but...

I just realised that the GFDL iceberg code is a part of SIS2 not MOM6 (e.g. see Martin and Adcroft 2010; Huth et al. 2022). I guess that means we won't have any means of including interactive icebergs in ACCESS-OM3. Does anyone know if NCAR/CESM folks have any plans to include icebergs in the future?

[anton-seaice]

Dave Bailey didn't identify it as a priority at the CESM Winter Working Group meeting - so its unlikely they are looking at it this year. I would guess they are still interested (although less relevant for Arctic work I guess).

We will have to do this at some point to couple with the ice-sheet model.

@adfraser is working on making a dataset of grounded icebergs to use in fast-ice modelling, maybe it could also be used for meltwater too.

[adfraser]

I did find this from CICE4 but it's pretty old by now - @anton-seaice do you know if it's been carried over to CICE6?
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2010JC006588

[anton-seaice]

I did find this from CICE4 but it's pretty old by now - @anton-seaice do you know if it's been carried over to CICE6? https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2010JC006588

Not that I know of, we could definitely ask Elizabeth though

[anton-seaice]

Till Rasmussen (DMO) mentioned these this morning:

https://opendrift.github.io/gallery/example_openberg_det.html

https://acciberg.nersc.no/

From a quick look they does post-processing of (iceberg and other) drift tracks etc. But I don't think it would take up volume in the sea-ice / melt etc ...

[adfraser]

Hi Anton,

That first one (opendrift) is pretty simple in terms of iceberg simulation - e.g. doesn't take sea surface slope into account which from memory is a big contributor to the drift. I could be wrong about this!

But @adele-morrison I think it could be good enough for that project we mentioned! NB this:
"To keep elements at coast until moved offshore at a later time (default for PelagicEgg):"
o.set_config('general:coastline_action', 'previous')
Sounds like grounded icebergs to me :)

I haven't look at the second one but if it's post-processing it could be useful to me too.

[adfraser]

Anton here are the iKID links I mentioned
https://www.gfdl.noaa.gov/modeling-tabular-icebergs/
and
https://www.gfdl.noaa.gov/iceberg-a68a/

edit: and the source code for that A68a simulation https://zenodo.org/records/6345238

[adfraser]

@ofa001 (did this tag work?) just told me that Alistair Adcroft has a pathway to porting iKID into CICE - Siobhan could you give more info on this?

[ofa001]

Hi @adfraser Its actually @sofarrell on the hive. It was during his COSIMA talk when he was talking about it and @AndyHoggANU asked questions of him, and he said it should be possible to treat the icebergs as a separate piece of code and link it to CICE. I did have a look at the code afterwards, but didnt follow up all the papers. Its definitely way more advanced than in earlier versions.
In relation to CICE iceberg developments it was on the agenda a while back as @anton-seaice says but then went quiet. I did ask Elizabeth about that, I think the person had left. I guess it would be good @anton-seaice to raise again at the Sante Fe meeting and see if they are going to integrate the GFDL icebergs.

[anton-seaice]

Someone else brought up at the consortium meeting today an no-one jumped to say yes they are pursuing it. I think E3SM might be interested (althought still not CICE).

[adfraser]

@AndyHoggANU: Adele and I are thinking of a PhD student in this space. If there was enough ACCESS support we could maybe consider getting the iKID model ported into CICE as their first chapter. Would need a GFDL and a CICE research advisor if we were getting serious about it, I think. This might not be insurmountable.
@ofa001 @anton-seaice @adele-morrison is this too ambitious?

[AndyHoggANU]

It does sound ambitious ... but it's something that I would like to see from an ACCESS point of view. So, we would be able to support on the software side provided there was someone able to take on the science side. Key thing would be to find the right candidate, and my guess is they are few and far between ....