[MITgcm-support] The criteria of cg3dMaxIters

[Jody Klymak]

On 15 Aug 2017, at 9:58, Saeid Esmaeilpour wrote:

> Thanks again Jody,
> It was really helpful, for the last question about this issue, Do I 
> need
> nonHydrostatic=.True. for the simulation of Eddies? Isn't that a 
> fine-scale
> phenomena? will MIT show up  Eddies at 2-degree resolution?

Do your eddies have large dw/dt?

Are they about 12-degrees across?

Cheers,  Jody

>
> Thanks,
>    Saeid
>
> On Tue, Aug 15, 2017 at 12:42 PM, Jody Klymak <jklymak at uvic.ca> wrote:
>
>> On 15 Aug 2017, at 9:29, Saeid Esmaeilpour wrote:
>>
>> Jody,
>> my model is about The simulation of currents Circulation from surface 
>> to
>> depth in indian Ocean, and I selected nonHydrostatic=.TRUE., cause I
>> think the model resolve the Nonlinear terms with nonHydrastatic=True 
>> (am I
>> wrong ?). If I select nonHydrastatic=True, in this case how much 
>> should I
>> set for cg3MaxIters? and If select nonHydrastatic=False, how much for 
>> that?
>> basically for the simulation of currents Circulation and at 2-degree 
>> I have
>> to take non Hydraustatic=false?
>>
>> nonHydrostatic=.True. solves the non-hydrostatic equation for dw/dt,
>> which essentially means the pressure needs to be determined globally 
>> from a
>> Poisson equation, and hence the 3-D solver. .False. means that 
>> pressure
>> is solved using the hydrostatic approximation (if you don’t know 
>> what this
>> is, Google plus a basic GFD course would help).
>>
>> This really has nothing to do with non-linearity until you start 
>> resolving
>> scales where the hydrostatic approximation breaks down (basically 
>> dw/dt is
>> no longer << rho’g / rho_0).
>>
>> You really only need non-hydrostatic approximation for fine-scale 
>> runs
>> looking at internal waves, dense overflows, convection, or other 
>> turbulent
>> phenomena. None of which will show up at 2-degree resolution.
>>
>> Cheers, Jody
>>
>> Thanks again,
>> Saeid
>>
>> On Tue, Aug 15, 2017 at 11:59 AM, Jody Klymak <jklymak at uvic.ca> 
>> wrote:
>>
>> Hi Saied,
>>
>> cg3dMaxIters is only used if nonHydrostatic=.True.. You aren’t 
>> resolving
>> any non-hydrostatic physics at 2-degree resolution, so I would 
>> suggest
>> setting nonHydrostatic=.False. Your model runs will be *much* faster.
>>
>>
>> Cheers, Jody
>>
>> On 15 Aug 2017, at 8:43, Saeid Esmaeilpour wrote:
>>
>> Hi Dear users,
>> could someone please explain to me what are the following PARM02?
>> when I increase cg3dMaxIters from 50 to 400 (e.g,. cg3dMaxIters=400) 
>> the
>> run speed becomes very slowly. how can I select this value? what's 
>> the
>> criteria for selecting this number? My domain is 510*210 with 20layer 
>> and
>> also my resolution is 2 degree (3704m).
>>
>> # Elliptic solver parameters
>> &PARM02
>> cg2dMaxIters=1000,
>> cg2dTargetResidual=1.E-13,
>> cg3dMaxIters=400,
>> cg3dTargetResidual=1.E-13,
>>
>> Thanks for any help,
>> Saeid
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>>
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>>
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>>
>>


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