[mpas-developers] /home/subversion/mpas/model revision 594

Michael Duda duda at ucar.edu
Thu Nov 4 19:11:45 MDT 2010


Hi, Todd.

Thanks for the link to Qingshan's paper.

I can add this same change to the sw and ocean cores in our
atmos_nonhydrostatic branch, and then, if there are no objections,
I can merge these changes back into the trunk. Is there a
preferred namelist record to add config_apvm_upwinding under in
the ocean core (I noticed that many options have been split out
into, e.g., hmix and vmix, which I think is nice)? Or perhaps a
better name for the option than config_apvm_upwinding?

Cheers,
Michael


On Thu, Nov 04, 2010 at 06:26:43PM -0600, Todd Ringler wrote:
> 
> Hi Micheal et al.,
> 
> Moving this to the namelist makes sense. Also, Qingshan has a  
> manuscript in review that I think is a significant improvement to the  
> APVM scheme (link is here: http://public.lanl.gov/ringler/files/ 
> apvm.pdf ). He will be implementing this into the shallow-water and  
> ocean models. If there is interest, we can consider putting this into  
> the frameworks to allow the different cores to have access to this  
> scale-aware version of APVM.
> 
> Cheers,
> Todd
> 
> On Nov 4, 2010, at 5:35 PM, mpas-developers at ucar.edu wrote:
> 
> >duda 2010-11-04 17:35:55 -0600 (Thu, 04 Nov 2010)
> >
> >BRANCH COMMIT
> >
> >Allow pv upwinding to be controlled by a namelist parameter,
> >config_apvm_upwinding, which defaults to 0.5.
> >
> >
> >M src/core_hyd_atmos/Registry
> >M src/core_hyd_atmos/module_time_integration.F
> >M src/core_nhyd_atmos/Registry
> >M src/core_nhyd_atmos/module_time_integration.F
> >Modified: branches/atmos_nonhydrostatic/src/core_hyd_atmos/Registry
> >===================================================================
> >--- branches/atmos_nonhydrostatic/src/core_hyd_atmos/Registry	 
> >2010-11-04 00:19:10 UTC (rev 593)
> >+++ branches/atmos_nonhydrostatic/src/core_hyd_atmos/Registry	 
> >2010-11-04 23:35:55 UTC (rev 594)
> >@@ -18,6 +18,7 @@
> > namelist logical   sw_model config_positive_definite    false
> > namelist logical   sw_model config_monotonic            true
> > namelist integer   sw_model config_mp_physics           0
> >+namelist real      sw_model config_apvm_upwinding       0.5
> > namelist integer   dimensions config_nvertlevels        26
> > namelist character io       config_input_name           grid.nc
> > namelist character io       config_output_name          output.nc
> >
> >Modified: branches/atmos_nonhydrostatic/src/core_hyd_atmos/ 
> >module_time_integration.F
> >===================================================================
> >--- branches/atmos_nonhydrostatic/src/core_hyd_atmos/ 
> >module_time_integration.F	2010-11-04 00:19:10 UTC (rev 593)
> >+++ branches/atmos_nonhydrostatic/src/core_hyd_atmos/ 
> >module_time_integration.F	2010-11-04 23:35:55 UTC (rev 594)
> >@@ -1988,7 +1988,7 @@
> >       !
> >       do iEdge = 1,nEdges
> >          do k = 1,nVertLevels
> >-           pv_edge(k,iEdge) = pv_edge(k,iEdge) - 0.5 * v(k,iEdge) *  
> >dt * gradPVt(k,iEdge)
> >+           pv_edge(k,iEdge) = pv_edge(k,iEdge) -  
> >config_apvm_upwinding * v(k,iEdge) * dt * gradPVt(k,iEdge)
> >          end do
> >       end do
> >
> >@@ -2027,7 +2027,7 @@
> >       !
> >      do iEdge = 1,nEdges
> >         do k = 1,nVertLevels
> >-          pv_edge(k,iEdge) = pv_edge(k,iEdge) - 0.5 * u(k,iEdge) *  
> >dt * gradPVn(k,iEdge)
> >+          pv_edge(k,iEdge) = pv_edge(k,iEdge) -  
> >config_apvm_upwinding * u(k,iEdge) * dt * gradPVn(k,iEdge)
> >         end do
> >      end do
> >
> >
> >Modified: branches/atmos_nonhydrostatic/src/core_nhyd_atmos/Registry
> >===================================================================
> >--- branches/atmos_nonhydrostatic/src/core_nhyd_atmos/Registry	 
> >2010-11-04 00:19:10 UTC (rev 593)
> >+++ branches/atmos_nonhydrostatic/src/core_nhyd_atmos/Registry	 
> >2010-11-04 23:35:55 UTC (rev 594)
> >@@ -30,6 +30,7 @@
> > namelist integer   nhyd_model config_mp_physics           0.
> > namelist real      nhyd_model config_epssm                0.1
> > namelist real      nhyd_model config_smdiv                0.1
> >+namelist real      nhyd_model config_apvm_upwinding       0.5
> > namelist integer   dimensions config_nvertlevels        26
> > namelist character io       config_input_name           grid.nc
> > namelist character io       config_output_name          output.nc
> >
> >Modified: branches/atmos_nonhydrostatic/src/core_nhyd_atmos/ 
> >module_time_integration.F
> >===================================================================
> >--- branches/atmos_nonhydrostatic/src/core_nhyd_atmos/ 
> >module_time_integration.F	2010-11-04 00:19:10 UTC (rev 593)
> >+++ branches/atmos_nonhydrostatic/src/core_nhyd_atmos/ 
> >module_time_integration.F	2010-11-04 23:35:55 UTC (rev 594)
> >@@ -3176,7 +3176,7 @@
> >       !
> >       do iEdge = 1,nEdges
> >          do k = 1,nVertLevels
> >-           pv_edge(k,iEdge) = pv_edge(k,iEdge) - 0.5 * v(k,iEdge) *  
> >dt * gradPVt(k,iEdge)
> >+            pv_edge(k,iEdge) = pv_edge(k,iEdge) -  
> >config_apvm_upwinding * v(k,iEdge) * dt * gradPVt(k,iEdge)
> >          end do
> >       end do
> >
> >@@ -3217,11 +3217,11 @@
> >
> >       ! Modify PV edge with upstream bias.
> >       !
> >-!     do iEdge = 1,nEdges
> >-!        do k = 1,nVertLevels
> >-!          pv_edge(k,iEdge) = pv_edge(k,iEdge) - 0.5 * u(k,iEdge)  
> >*dt * gradPVn(k,iEdge)
> >-!        end do
> >-!     end do
> >+      do iEdge = 1,nEdges
> >+         do k = 1,nVertLevels
> >+            pv_edge(k,iEdge) = pv_edge(k,iEdge) -  
> >config_apvm_upwinding * u(k,iEdge) *dt * gradPVn(k,iEdge)
> >+         end do
> >+      end do
> >
> >
> >    end subroutine compute_solve_diagnostics
> >
> >
> >_______________________________________________
> >mpas-developers mailing list
> >mpas-developers at mailman.ucar.edu
> >http://mailman.ucar.edu/mailman/listinfo/mpas-developers
> 

> _______________________________________________
> 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