[MITgcm-support] Bottom Pressure nonhydrostatic

[Martin Losch]

David,

this is a tricky question, and I thought that I knew the answer, but  
after having a closer look at the code I realized that I no longer  
understand this business, but I know this much:
in z-coordinates PHL, is the the pressure anomaly divided by rhoConst  
at the bottom of the domain. It's a pure diagnostics, and it's  
computed in model/src/diags_phi_rlow.F
It's an (potential) anomaly and to get the total hydrostatic phiHyd,  
you need to add that contant part gravity*abs(z).
Then in some cases you need to add the surface contribution as in  
totPhiHyd (diags_phi_hyd.F:  Bo_surf*etaN+phi0surf, Bo_surf = gravity  
in z-coordinates, see ini_linear_phisurf.F), but I ran across a place  
(calc_phi_hyd.F) where (for a nonlinear free surface and if  
select_rStar.EQ.0 .AND. nonlinFreeSurf.GT.3)  
surfPhiFac*etaH(i,j,bi,bj) is added to the surface value of phiHyd).  
That's most likely correct, but I don't know why. Maybe Jean-Michel  
has a simle explanation (probably has to do with time stepping and the  
pressure being updated at a different time than etaH and etaN).

The non-hydrostatic pressure is completely separate and is never part  
of PHL nore phiHyd (which as the name implies is completely  
hydrostatic).

Martin

On Oct 14, 2009, at 6:35 PM, David Hebert wrote:

> In grepping the code (DYNVARS.h specifically) it looks like PHL is  
> the bottom hydrostatic pressure. Two question about this..
>
> 1) Does this include surface contribution  (as specified for  
> totPhiHyd in DYNVARS.h)
> 2) In a nonhydrostatic simulation would I obtain bottom pressure as
> PHL + PNH(bottom)?
> Or do I need to use
> PH(bottom) + PNH(bottom)?
>
> Thanks,
>
> David
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support