GOSI11: Southern Ocean

[oceanaguiar]

Ongoing work to improve simulation in the Southern Ocean.

• Southern Ocean WG
• Submit paper on Southern Ocean biases and PEG work. DS
• Work on new bathymetries (see #1 ). DS
• Investigate deep water formation and evolution of AABW. DS and others
• Deep water formation diagnostic (VALSO metric?).
• Age tracer to look at cascades?
• AABW census metric?
• Diagnose spurious mixing? (Or change in isopycnal height in spin up)
• Look at impact of embedded MEs coordinates and/or AGRIF on simulation – especially on deep water formation and evolution. CG
• Plan for GOSI + ice cavities (currently only active in UKESM, MEs in the future?)
• Ongoing work on icebergs (UKESM/BAS/MetO).

Wish list/other possibilities:

• Investigate the possibility of parametrising sub-gridscale bathymetric roughness via spatially varying bottom friction.

[atb299]

• Deep water formation diagnostic (VALSO metric?)

For ORCHESTRA George and I looked at a pair of short simulations comparing the Z and SZT vertical coordinates. George calculated AABW volume, showing that the SZT model preserved the volume much better than Z did. It might be a useful metric for this. Details are in here: https://royalsocietypublishing.org/doi/full/10.1098/rsta.2022.0070 (see end of section 3c and figure S4).

[ukmo-cguiavarch]

Testing the impact of MEs in the Southern Ocean

Preliminary work before investing more time in setting up MEs specifically targeting the Antarctic shelves. Test the impact of MEs along Antarctica in eORCA025 configuration using the MEs already computed for SE-NEMO, where MEs setup was optimised for the global shelves, not specifically for Antarctica -> when we will develop the MEs setup specifically for the Southern Ocean, we will have more freedom to explore more advanced/extended configurations.

• [x] creating domain cfg with MEs along the Antarctic shelves using the MEs set up for SE-NEMO #10 ( @oceandie ) : /data/d03/dbruciaf/NEMO/SO/orca025/domcfg/domain_cfg_ant_se-nemo_like.nc

• [x] setting up and running suite with the same set up used for the Nordic Overflow (JRA55 forcing, EN4 initialisation,(https://github.com/JMMP-Group/GO_coordination/issues/15#issuecomment-1840425451) ( @ukmo-cguiavarch ): u-dc834 - integration completed: 45y forced by JRA initailised from EN4

• [ ] validation ( @ukmo-cguiavarch,@oceandie, Jeff R )

  • VALSO metrics for GOSI9, GOSI9 (triads + bot fric) and GOSI9 MEs SO

[ukmo-cguiavarch]

Impact of resolving ice shelf cavities on the Southern Ocean

• VALSO metrics for GO6 and GO7 highlights the impact of resolving the ice cavities along Antarctica on the transports and on the T/S properties on the shelves. With the resolved ice cavities in GO7, the ACC transport is increased to the same level as GOSI9 but in GO7, it is associated with significant improvements in the T/S properties on the shelves (this is not the case in GOSI9).

• Note that in GO7, the ice cavities are opened but it does not use the interactive ice shelf melt (cn_isfcav_mlt = '2eq/3eq' in NEMO4.2) and instead uses prescribed ice melt (cn_isfcav_mlt = 'spe' in NEMO4.2). It uses the same ice shelf melt climatology than GO6 (Rignot et al, 2013) but uses the mapping from a regional run with interactive iceshelf melt from Mathiot et al (2017).