[MITgcm-support] Kinetics energy budget

[Jean-Michel Campin]

Hi Marco,

Can you clarify this:
> high values for the coriolis term , that is impossible

I don't know the type of flow regime you are working with (this
could be useful to know) but, in many MITgcm applications,
the flow is not very far from geostrophy and the Coriolis term
is one of the 2 major terms. Given the C-grid discretisation,
I don't expect a local (i.e., at each grig point) cancellation 
of V*(f*U) and U*(f*V) terms, but rather only approximately on a wider stencil.

Also, might be useful to be sure that you are able to close the momentum
budget (for each U,V component), and there has been several post 
to this list regarding this subject. The point here is that this should 
translate into a quite accurate KE budget (down to time-stepping error,
i.e.  adams-bashforth).

And finally, anytime I have to check KE budget, I tend to start from 
the KE expression in monitor/mon_ke.F (we really should make a proper 
"diagnostics" from it), which is probably better than 
pkg/mom_common/mom_calc_ke.F (with KEscheme=2).

Cheers,
Jean-Michel

On Mon, Oct 26, 2015 at 01:59:22PM +0100, marco reale wrote:
> hello MITGCM users,
> 
> I was wondering if it is possible using MITGCM momentum budget diagnostics estimating the kinetics energy budget of a domain.
> 
> I mean given:
> 
> Um_coriolis+Um_diss+Um_dPhDx+SSHgradient+Um_Adv+Um_ext , I multiplied for Uvel and I do the the same for the meridional velocity component.
> 
>  After if I apply the definition of KE related to the scheme 2 (the default for flux form) I should reconstruct the variation of kinetics energy???But I found analyzing each term for example high values for the coriolis term , that is impossible 
> 
> have anyone among you already tried this kind of computation?
> 
> cheers
> 
> marco
> 
> 
> 
> 
> 
> 
> 
> 

> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support