[Mitgcm-support] find_rho question

[mitgcm-support at dev.mitgcm.org]

Hi.

  Apologies in advance if the current algorithm is correct.

The LINEAR eos portion of find_rho (v. 1.15) reads as follows:

C ***NOTE***
C In the linear EOS, to make the static stability calculation meaningful
C we alway calculate the perturbation with respect to the surface level.
C **********
       refTemp=tRef(kRef)
       refSalt=sRef(kRef)

       do j=jMin,jMax
        do i=iMin,iMax
         rholoc(i,j)=rhonil*(
     &     sBeta*(sFld(i,j,k,bi,bj)-refSalt)
     &   -tAlpha*(tFld(i,j,k,bi,bj)-refTemp) )
        enddo
       enddo

Should the reference temperature and salinity be 

       refTemp=tRef(1)
       refSalt=sRef(1)

to make the resulting density independent of the given reference depth (and be
consistent with the NOTE)?  

With a linear eos, I would think changing the reference depth (kRef) should not
change the returned rholoc.  In the current algorithm, if initial reference
T and S are depth-dependent, then find_rho will return different densities for
the same tFld and sFld, but different kRefs.

Looking through the CVS tree, I see that the referencing was changed from 1 to
kRef in version 1.6, and the change noted explicitly.  Versions 1.5 and 1.4 had
it as I am suggesting.  The fact that it was changed suggests that I'm probably
mistaken in my logic, but a quick check that the current form is correct would
be great.

Thanks.
                                                        -Chris
       
---------------------------------------------------------------------
Christopher Edwards                     christopher.edwards at uconn.edu
Department of Marine Sciences           phone: (860) 405-9157
University of Connecticut               fax: (860) 405-9153
1080 Shennecossett Road
Groton, CT  06340-6048
---------------------------------------------------------------------