Hello,
There is a problem in the kpp code when when kmt=2. The
following change in ocean_vert_mix_coeff.F90 (in
mom4/ocean_param/mixing/vert/kpp/ocean_vert_mix_coeff.F90)
fixes the problem. Replace
lines 974-986 in the original code with the following:
!-----------------------------------------------------------------------
! find new kbl
!-----------------------------------------------------------------------
! For kmt > 2
do kl=2,nk
do j=jscomp,jecomp
do i = iscomp,iecomp
if((kbl(i,j) == kmt(i,j)).and.(zt(kl) > hbl(i,j)) .and.
(kmt(i,j) > 2)) then
kbl(i,j) = kl
endif
enddo
enddo
enddo
! Fix when kmt =2
do kl=1,2
do j=jscomp,jecomp
do i = iscomp,iecomp
if((kbl(i,j) == kmt(i,j)).and.(zt(kl) > hbl(i,j)) .and.
(kmt(i,j) == 2)) then
kbl(i,j) = kl
endif
enddo
enddo
enddo
-----------------------------------------------------------------------
With best regards
Swathi
PS: I have also attached a mail that Martin Schmidt sent me when I was
trying to locate the problem. His mail has more general applicabilty as
far as the way heat flux and radiation are handled within kpp. If you have
comments on this, please send your view to the mom4 bulletin board.
Thanks.
On Fri, 8 Aug 2003, Martin Schmidt wrote:
> Hi,
>
> I had a similar problem with kpp and MOM-31. Temperature became
> negative at some grid points. This is not noticed by the
> equation of state and the run went through. The effect was not
> to be seen in the output, since only time means had been
> written, but could be detected from snapshots. The point was,
> that the downward heatflux, entering the kpp-scheme of MOM-3
> includes the total downward short wave radiation. At the same
> time this part of the heat flux is distributed separately over
> the water column by an exponential absorption model. This
> contradicts the ideas of Large et al. where the surface tracer
> flux is redistributed by tubulence down to some pycnocline even
> if there is no tracer gradient. In the case of heat, this must
> be the energy, which has just been added to the enthalpy of the
> water, but not the radiation going downward below the surface
> level. My opinion is, that only the radiation absorbed in the
> uppermost level should be added to the surface heatflux. I have
> modified kpp and tracer in this sense which removes the problem.
>
> I was away for several weeks and I am back the first day. So I
> cannot look into the kpp-code of MOM4 today. Anyway, may be this
> e-mail helps you, to fix the problem. If you want, I send you my
> version of tracer.F of MOM-31 with the new algorithm.
>
> Regards,
> Martin
>
-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01
_______________________________________________
FMS-mom4 mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fms-mom4
