[MITgcm-support] seaice drag

[Martin Losch]

Hello again,

I have a (maybe stupid) question about the drag computations in pkg/ 
seaice/dynsolver.F, maybe someone can comment:

starting at line 112, the stress is computed as follows (I am  
ommiting the loop statements)
> C--   Wind stress is computed on South-West B-grid U/V
> C     locations from wind on tracer locations
>           U1=QUART*(UWIND(I-1,J-1,bi,bj)+UWIND(I-1,J,bi,bj)
>      &             +UWIND(I  ,J-1,bi,bj)+UWIND(I  ,J,bi,bj))
>           V1=QUART*(VWIND(I-1,J-1,bi,bj)+VWIND(I-1,J,bi,bj)
>      &             +VWIND(I  ,J-1,bi,bj)+VWIND(I  ,J,bi,bj))
>           AAA=U1**2+V1**2
>           IF ( AAA .LE. SEAICE_EPS_SQ ) THEN
>              AAA=SEAICE_EPS
>           ELSE
>              AAA=SQRT(AAA)
>           ENDIF
> C first ocean surface stress
>           DAIRN(I,J,bi,bj)=RHOAIR*OCEAN_drag
>      &         *(2.70 _d 0+0.142 _d 0*AAA+0.0764 _d 0*AAA*AAA)
>           WINDX(I,J,bi,bj)=DAIRN(I,J,bi,bj)*(COSWIN*U1-SINWIN*V1)
>           WINDY(I,J,bi,bj)=DAIRN(I,J,bi,bj)*(SINWIN*U1+COSWIN*V1)
>
> C now ice surface stress
>           DAIRN(I,J,bi,bj)=RHOAIR*(SEAICE_drag*AAA*AREA(I,J,1,bi,bj)
>      &         +OCEAN_drag*(2.70 _d 0+0.142 _d 0*AAA
>      &         +0.0764 _d 0*AAA*AAA)*(ONE-AREA(I,J,1,bi,bj)))
>           FORCEX(I,J,bi,bj)=DAIRN(I,J,bi,bj)*(COSWIN*U1-SINWIN*V1)
>           FORCEY(I,J,bi,bj)=DAIRN(I,J,bi,bj)*(SINWIN*U1+COSWIN*V1)

The way I understand this is that the stress on the ice is an average  
over ocean stress and ice stress ( the statement before FORCEX 
(I,J,bi,bj) = ...), whereas the stress over the ocean is just the  
ocean stress not weighted at all. I find this a little inconsistent  
if not wrong. I would think that these terms should be treated  
separately, with only the ice stress driving the ice, right?

Martin