Surface freshwater fluxes in 1m and 5m simulations

[willaguiar]

In the review process, one of the requests was for a freshwater budget to see the effects of salinity restoring, sea ice formation and freshwater advection on the shelf salinity. While we cannot do that without fully rerunning simulations, we can still gather the effect of restoring and sea ice by just analyzing the surface freshwater fluxes over the shelf.

---

Below is the total freshwater fluxes on the shelf for the simulations (+ = water into ocean). The fluxes below are integrated over the continental shelf, and averaged over the first 6 months of each simulation (which is the timescale where salinity changes over the shelf)

These fluxes confirm that salinity restoring is just too small to matter.
A reviewer might be caught up in the difference of sea ice fluxes between the 5m and 1m (differ by ~ 4% in sea ice ratio, and 11% in total FW ratio). I guess we can also point for the results of figure 4 (and the new quantification we did for the effect of sea ice on the text) as a second proof that the change in sea ice fluxes has a low effect on DSW formation.

Bias warning: I didn't have specific outputs for P-E+R for MOM6, or FWF specifically from sea ice. But since the forcing is the same, I just used the P-E+R from MOM5, and calculate the sea ice fluxes in MOM6 as the residual

[adele-morrison]

I think this is good to include. But... we can't use P-E+R from MOM5 in MOM6, because E changes depending on the ocean state. If it's just 6 months, can we rerun to output P, E, R directly from MOM6?

[willaguiar]

Here is an update using the precipitation, evaporation and runoff outputs from the MOM6 reruns .The decrease ratios are pretty much the same (differ by ~ 5% in sea ice ratio, and 11% in total FW ratio). MOM6 seem to have more precipitation in these first 6 months, which I think it is because of the zero-sea ice initial conditions. Im updating this fig to add to the suppmat

[willaguiar]

One additional info that might be useful for setting up the diag_table in future MOM6 runs. None of the output fields from MOM6 ocean (i.e. ocean_month) have an estimate of freshwater fluxes exclusively from sea ice (i.e., seaice_melt=frunoff=fsitherm=0). Instead, the total net fluxes to the ocean (wfo) match the sum of precip + evap + lrunoff, with lrunoff = 0.08 Sv being a constant field for our prescribed value from Depoorter. So sea ice melt and freezing seem to be within precip + evap. To get the sea ice fluxes here I looked at the net_melt field from SIS output (ice_month), which contains sea ice formation+melt+snowfall, and then removed the snowfall flux (fprec) from it. So it might be useful to have fprec and net_melt set up as defaults in the diag_table.
#54

[adele-morrison]

One additional info that might be useful for setting up the diag_table in future MOM6 runs.

Good idea! Want to edit the main panan diag_table on github to include different surface FW flux terms? Did you also check with Christina? Pretty sure she's looked at sea ice fluxes before.