Salinity/Temp Anomaly/initialisation bug

[claireyung]

The salinity anomaly still persists with non-Boussinesq ice shelf. See this notebook https://github.com/pedrocol/MOM6_pge_tests/blob/claire/notebooks/Plot_IC.ipynb

Salinity anomaly is very close to the seamount. Unlike the sigma_shelf_zstar case I was trying with triangular ice shelf, it is only very close to the vanishing layers (probably due to the high res, note only 5th layer interface shown below, and the zstar coordinate), and therefore velocities get damped out quickly unlike my triangular ice shelf, leading to lower velocities.

[adele-morrison]

Just to clarify: this anomaly is the initialisation bug?

[adele-morrison]

As discussed today: Try to fix the initialisation with Angus' offline python thing to call the relevant MOM routine.

Also, as summarised today: the initialisation hack works in 2D, but not in 3D cases. It's not clear why it doesn't work in 3D.