[MITgcm-devel] seaice advection schemes

[Jean-Michel Campin]

Hi Martin,

Did not see any answer to your message.
And mine might not be very useful.
I think DST3 is 3rd order in space, but I don't know if it's also 3rd
order in time (never tried to check this).
Might be easy to check in simple advection test (with constant in time flow),
better if we know the analytic solution (e.g., 1-D advection test, then 2-D 
test such as advect_xy, advect_cs).
Or did you already tried the 1-D advection test (with uniform & constant uVel) ?

Cheers,
Jean-Michel

On Fri, Jun 21, 2013 at 01:36:07PM +0200, Martin Losch wrote:
> Hi there,
> 
> I am trying to build a 2nd-order scheme for sea-ice time stepping, but I can't. I only get 1st order convergence, and I think the culprit is the advection of seaice (with seaice_advection). I am assuming that schemes like DST3 are 3rd order in space AND time, but I cannot show this experimentally (with constant uice in 1D channel I still only get 1st order convergence). 
> 
> Is my assumption about the order DST3 (and other scheme in seaice_advection) wrong?
> 
> Martin
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel