Make sure we keep salinity restoring constant as we change gridcell thickness

[adele-morrison]

From the ACCESS-OM2-01 development doc: "The restoring timescale is set by salt_restore_tscale; we use 10 days at 0.1deg and 21.28 days for the coarser resolutions. The “piston velocity“ (surface vertical grid spacing divided by restoring timescale) is more physically relevant than the restoring timescale. This is 0.11 m/day at all resolutions due to the differing vertical resolution (see Table 4). For comparison, 15 m/300 days (i.e. 0.05 m/day) is GFDL’s standard piston velocity (Griffies, 12 April 2018), so we are restoring about twice as strongly as GFDL."

Make sure we adjust this when we change the top gridcell thickness.

[willaguiar]

Thanks! I just adjusted salt_restore_tscale to 45.4 days

[willaguiar]

By now, I have set-up the the MOM6 (panan01-5mtop) case, with increased thickness of the top cell to 5m, but we need to figure out how to deal with the salinity restoring adjustment. For the ACCESS-OM2 case, we did not change the piston velocity ( 0.11 m/day) when coarsening the vertical resolution - instead, we changed the salinity restoring timescale from 10 days to 45.4 days, proportional to the top cell thickness increase to 5 meters. However, I looked up in _MOM_parameterdoc.all,and MOM6 do not seem to have an option to change the restoring timescale. Do you reckon we should just change the piston velocity in panan01-5mtop the instead? ( Currently 0.1667 m/day) Or does anyone have additional suggestions?

[adele-morrison]

We want the piston velocity to be the same in all experiments. Perhaps MOM6 adjusts the restoring scale automatically depending on vertical resolution? Maybe worth a check in the code to see?

[adele-morrison]

Andy and Andrew reckon that MOM6 automatically adjust this for us. Just leave the piston velocity constant across all experiments.

[willaguiar]

Awesome, will do!