[MITgcm-devel] Held Suarez

[chris hill]

Jean-Michel,

 I'm updating the Held-Suarez write up.
 Any thoughts on why we don't use rC in the formula below for termP?
 e.g. termP = MIN(rC(kLev)*recip_P0g,1.)
 is it some NLFS thing?

Chris

      DO J=1,sNy
       DO I=1,sNx+1
        IF ( hFacW(i,j,kLev,bi,bj) .GT. 0. ) THEN
         recip_P0g=MAX(recip_Rcol(I,J,bi,bj),recip_Rcol(I-1,J,bi,bj))
         termP=0.5 _d 0*( MIN(rF(kLev)*recip_P0g,1. _d 0)
     &                   +rF(kLev+1)*recip_P0g )
         kV=kF*MAX( 0. _d 0, (termP-sigma_b)/(1. _d 0-sigma_b) )
         gU(i,j,kLev,bi,bj)=gU(i,j,kLev,bi,bj)
     &                      -kV*uVel(i,j,kLev,bi,bj)
        ENDIF
       ENDDO
      ENDDO