No subject

formulation, a salt tendency is added to the two grid cells vertically
abutting the ice shelf, with the one closest to the ice shelf being denoted
with kLev=kTopC.  Then skipping to the end,

gSloc(i,j,kTopC, bi,bj)=shelficeForcingS(i,j,bi,bj)  / (
drF(kTopC)*hFacC(kTopC) + drLoc)

and

gSloc(i,j, kTopC+1 , bi,bj)= shelficeForcingS(i,j,bi,bj) /
(drF(kTopC)*hFacC(kTopC) + drLoc)

where drLoc is the same for both = drF(kTopC)*(1-hFacC(i,j,kTopC, bi,bj)).

and gS=gS+gSLoc

It looks like both gSloc terms have the same numerator and denominator, but
I'm pretty sure the denominator isn't always 2.  Then the total salt
tendency actually added to the column (i,j) is not equal to
shelficeForcingS, which we have been assuming for our 2-d diagnostic.

So did we assign the diagnostic wrong or are those terms supposed to have
different denominators or have I missed something?

Thanks,

Madeline

--00163630f445bf9bc004860a410c
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Dear Martin/anyone familiar with the shelfice package,<br><br>A while back =
we added a diagnostic for the salt forcing from the ice shelf, SHIForcS<br>=
where SHIForcS =3D shelficeForcingS*rUnit2mass (g/m^2/s).<br><br>I am addin=
g a 3d tracer for the quantities added to gS in the &quot;boundary layer&qu=
ot; and it looks like there is a discrepancy between our diagnostic and wha=
t is actually being added to gS , but I am probably just confused so I woul=
d be very happy if you could offer some clarification on the salt addition =
scheme. <br>

<br>From what I understand in shelfice_forcing.F, using the boundary layer =
formulation, a salt tendency is added to the two grid cells vertically abut=
ting the ice shelf, with the one closest to the ice shelf being denoted wit=
h kLev=3DkTopC.=A0 Then skipping to the end, <br>

<br>gSloc(i,j,kTopC, bi,bj)=3DshelficeForcingS(i,j,bi,bj)=A0 / ( drF(kTopC)=
*hFacC(kTopC) + drLoc)<br><br>and <br><br>gSloc(i,j, kTopC+1 , bi,bj)=3D sh=
elficeForcingS(i,j,bi,bj) / (drF(kTopC)*hFacC(kTopC) + drLoc)<br><br>where =
drLoc is the same for both =3D drF(kTopC)*(1-hFacC(i,j,kTopC, bi,bj)).=A0 <=
br>

<br>and gS=3DgS+gSLoc<br><br>It looks like both gSloc terms have the same n=
umerator and denominator, but I&#39;m pretty sure the denominator isn&#39;t=
 always 2.=A0 Then the total salt tendency actually added to the column (i,=
j) is not equal to shelficeForcingS, which we have been assuming for our 2-=
d diagnostic.=A0 <br>

<br>So did we assign the diagnostic wrong or are those terms supposed to ha=
ve different denominators or have I missed something?<br><br>Thanks,<br><br=
>Madeline<br>

--00163630f445bf9bc004860a410c--