[Mitgcm-support] Short Wave Flux divergence

[mitgcm-support at dev.mitgcm.org]

In the routine external_forcing_t.F the divergence of the penetrating
short-wave flux (Qsw) is using a constant thickness - delz(1) - rather
than the local thickness - delz(k).  This makes the heating stronger
at depth and the depth integral does not add up to the surface flux.
This expression has been this way since the very beginning
even when it was first added in calc_gt.F (version 1.20).

> #ifdef SHORTWAVE_HEATING
> C Penetrating SW radiation
>       swfracb(1)=abs(rF(klev))
>       swfracb(2)=abs(rF(klev+1))
>       call SWFRAC(
>      I     two,minusone,
>      I     myCurrentTime,myThid,
>      O     swfracb)
>       DO j=jMin,jMax
>        DO i=iMin,iMax
>         gT(i,j,klev,bi,bj) = gT(i,j,klev,bi,bj)
>      &    -maskC(i,j)*Qsw(i,j,bi,bj)*(swfracb(1)-swfracb(2))
>      &    *recip_Cp*recip_rhoNil*recip_dRf(1)
>        ENDDO
>       ENDDO
> #endif

We are wondering if this is correct or whether this is no longer used at all?

Jean-Michel & Alistair.