[MITgcm-devel] more seaice

[Martin Losch]

Hi,

independently from all of the previous emails about seaice I would like 
to make a suggestion for the seaice-package: I would like to introduce 
masks (maskC) into the routine GROWTH for:
---> surfaceTendencyTice (line 139), because of neatness
---> theta (after line 139), because of the substraction of two "large" 
numbers to prevent non-zero values over land
---> EmPmR (line 287), because the nonzero emp over land will lead to 
nonzero sealevel over land under certain circumstances (e.g. if 
convertFW2Salt = 35, or useRealFreshWaterFlux, or other up-to-now 
unphysical situtations). I am not sure whether this is a real problem, 
but it affects the values of cg2d_b, which are not masked. Over time, 
however, Eta can become very large over land (60m after 10yrs, in my 
case). I don't know if it makes sense to have a nonzero EmPmR over 
land? Certainly not if it's computed from a SEA ice model (o:
(---> Qnet is another candidate, but that has not lead to any visible 
problems.)

Martin