Legacy wave mixing & Stokes forces terms in MOM6

[gustavo-marques]

I am resurrecting an old email chain about assessing legacy wave mixing in MOM6 and the initial implementation of the wave-averaged Boussinesq equation in MOM6 (passing 3 stokes drift components from WW3 to MOM6).

From @alperaltuntas:
"I have recently started three JRA-forced CESM-MOM6 experiments to assess the impacts of wave coupling and Stokes forces terms in MOM6. All three runs span the JRA years 30-61, (1987-2018), and they are all close to being complete. MOM6 and WW3 are run on the same horizontal grid (0.66 degree tripolar). The cases are:
no waves: Out-of-the-box CESM-MOM6-CICE configuration
legacy: CESM's legacy wave coupling implemented between MOM6 and WW3. This coupling method was originally implemented between POP2 and WW3. The only difference with (1) is the enhanced turbulent velocity scale. Enhancement factor is received from WW3.
surfbands: Our new MOM6-WW3 coupling based on passing 3 stokes drift components from WW3 to MOM6. All three stokes forced terms are turned on within MOM6.

Please find below a link to a notebook with several surface fields, global fields, and stokes forced terms plots. You'll need to scroll down a bit to skip some python code. The diagnostics are either 20-year means (30-49) or 5 year means (45-49) as noted in the titles. Let me know if you have any questions.

https://github.com/alperaltuntas/notebooks/blob/main/diags/mom6/ww3/control_220309.ipynb

The summary of my takeaways are:

• As we expected (and hoped), surfbands case looks very similar to the no waves case, whereas the legacy case shows some divergence when compared to no waves. So the stokes forced terms don't appear to impose drastic changes.
• The biggest difference between the no waves case and the legacy case is in the boundary layer depth. In the Southern ocean, the turbulent velocity scale enhancement deepens the boundary layer by about 10m in summer and by about 30m in the winter. This is similar to the response that we get in POP-WW3 runs when the turbulent velocity scale enhancement is on (by default.)
• At the very end of the notebook, I plot 5-year means of stokes forces terms next to their counterparts. Both the stokes pressure forces and the stokes vortex forces are two orders of magnitude smaller. However, the stokes velocity tendency is about an order of magnitude smaller than the total velocity tendency. I'll let Brandon comment whether these results are assuring. "