[MITgcm-support] CS64 grid

[Martin Losch]

Dimitris, if you want an accurate solution to the momentum equations, the 1-day time step is not very good (let alone the 2 default pseudo-timesteps, see Lemieux and Tremblay, 2009, for more details), but  I don’t think that the 1-day time step is the cause for very thick ice. If 

I think that the LGM forcing leads to the very thick ice.

But I can be convinced otherwise, so setting SEAICE_deltaTdyn to some smaller value as a test should provide some insight.

Anthony, setting SEAICE_deltaTdyn = 1200, means that the ice solution will be totally wrong conceptually: Instead of stepping the ice model forward for a day, you just step it forward for 1200sec each time step, so you’ll slow down the development of ice dynamics by a factor of 24*3=72. But we can assume that the accelaration term in the ice-momentum equation is probably small (and negligible), so you should be OK with that setting. Please let us know what you find.

Martin

On Feb 10, 2014, at 5:18 PM, Dimitris Menemenlis <dmenemenlis at gmail.com> wrote:

> Martin, so 1-day time steps is OK for LSR?
> Wouldn't it be safer to set SEAICE_deltaTdyn = 1200.,
> consistent with deltaTmom instead of the default
> SEAICE_deltaTdyn = deltaTtracer = 86400., ?
> 
> On Feb 10, 2014, at 8:09 AM, Martin Losch <Martin.Losch at awi.de> wrote:
> 
>> The time step should be OK for this configuration.
>> 
>> What about your forcing? Is it too cold? Not enough incoming radiation? What do expect when you apply your forcing?
>> How long do you need to integrate before this happen?
>> 
>> Ultrathick ice appears usually in “corners” or “embayments” that have little connection to the rest of the domain, geometrically or dynamically, so that ice get’s stuck and keeps growing. For these cases capping in a thermodynamically inconsistent way is a convenient way out (HEFF=MIN(HEFF,someMaxValue), but that’s not implementend anymore, is it?
>> 
>> Martin
>> 
>> On Feb 10, 2014, at 4:59 PM, Anthony Coletti <ajcolett at geo.umass.edu> wrote:
>> 
>>> Is the sea ice dynamics time step located in data file or data.seaice?
>>> 
>>> These are the time steps in data file:
>>> 
>>> deltaTmom = 1200.,
>>> deltaTtracer = 86400.,
>>> deltaTfreesurf = 86400.,
>>> deltaTclock = 86400.,
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support