[mpas-developers] a bug in the SW code in the trunk

Mark Petersen mpetersen at lanl.gov
Fri Dec 9 13:29:00 MST 2011


Yes, I fixed that bug in the ocean core several months ago, and forgot 
to check for it in the sw core.  Thanks for finding it.  I did change 
the halo update to provis in the ocean core:

            if (config_h_mom_eddy_visc4 > 0.0) then
               call mpas_dmpar_exch_halo_field2d_real(domain % dminfo, 
provis % divergence % array(:,:), &
                                                block % mesh % 
nVertLevels, block % mesh % nCells, &
                                                block % parinfo % 
cellsToSend, block % parinfo % cellsToRecv)
               call mpas_dmpar_exch_halo_field2d_real(domain % dminfo, 
provis % vorticity % array(:,:), &
                                                block % mesh % 
nVertLevels, block % mesh % nVertices, &
                                                block % parinfo % 
verticesToSend, block % parinfo % verticesToRecv)
            end if

It should look the same in the sw core.  If you want to change it, that 
would be great.


On 12/09/11 13:12, Chen, Qingshan wrote:
> Hi everyone,
> I think I have found a bug in the SW code in the trunk. On lines 126 and 129 of mpas_sw_time_integration.F, the halo update for divergence and vorticity for computation of del4 diffusion is done for the time_level 2. However it only makes sense to update these two variables for provis, since it is provis that is used to compute the tendencies.
> This part of code was copied, most likely by me, from the ocean core. I have checked and seen that this bug has been fixed in the ocean core. If everyone agrees with me, then I would like to take the responsibility and fix this bug in the SW core. Since it is a very simple fix, I will likely just do it in the trunk and then commit the changes.
> Thanks!
> Qingshan
> _______________________________________________
> mpas-developers mailing list
> mpas-developers at mailman.ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/mpas-developers

More information about the mpas-developers mailing list