Hello,
We have NEMO4.2 coupled to our in-house atmosphere through our in-house coupler. (CanAM, CanCPL)
We are having some issues getting a realistic sea ice cover in coupled simulations, (our sea ice volume explodes 20 fold in 100 years) while forced simulations are giving a fairly good sea ice cover.
So we are investigating the fields going through our coupler and compare them to ones from the forced simulation (ERA5) through the NEMO bulk scheme “sbcblk”.
I noticed that the evap_ice field from our coupler is positive and negative while in the forced simulation, it is only positive. I then noticed the sentence in the SI3 documentation saying that
Sublimation converts the evaporation mass flux evap_ice into a snow loss
assuming constant density. The energy cost of sublimation is included in the latent
heat flux and needs not being accounted for. Evaporation removes snow mass, and
is handled layer by layer. Deposition of snow if any should be routed through the
solid precipitation mass flux. (p54, sect 6.4.3)
So I assume that we would move all deposition (negative evap_ice) to the sprecip while keeping the sublimation (positive evap_ice) in `evap_ice in our coupler
My issue here is that sprecip is also for fluxes going into the ocean, so I am not sure how to proceed here because it is realistic for snowfall but not for deposition, we should not deposit ice onto the ocean. We use the conservative setting for sn_rcv_emp in namsbc_cpl in namelist_cfg.
Also some comments in the code would suggest that deposition could be handled by SI3 through evap_ice? Has anyone experimented with this?
Thanks in advance!