[MITgcm-support] Spinup in a 2-D channel surface waves?

[Martin Losch]

Hi Jody,
If you want constant grid spacing you can use dxSpacing/dySpacing  
instead of delX/Y, in which case you would set
dxSpacing=200.,
regardless of your sNx. It would have avoided your little problem.

Martin

On 6 Oct 2006, at 00:23, Jody Klymak wrote:

>
> Hello,
>
> Oops, it was equally silly:
>
> delX = 200., is *not* the same as
> delX = 100*200.,
>
> It is dx = [200 99*1.234567e5].  That explains a lot...
>
> Sorry for the chatter.
>
> Cheers,  Jody
>
> On Oct 5, 2006, at 2:07 PM, Jody Klymak wrote:
>
>>
>> Hi Dmitri,
>>
>> Yes, the first and last depths in X are zero.
>>
>> I should mention that I have only one grid point in y.  Is that  
>> OK, or could it cause the surface algorithm to have difficulties?
>>
>> Thanks,  Jody
>>
>> On Oct 5, 2006, at 1:56 PM, Dmitri Leonov wrote:
>>
>>> Does your topography go all the way to the surface? (Otherwise  
>>> your channel will be periodic in the cross-channel direction).
>>> That's the only hint I can think of right now.
>>>
>>> Dmitri
>>>
>>> Jody Klymak wrote:
>>>>
>>>> Hello,
>>>>
>>>> I want to perform spinup of a 2-D channel to along-channel winds  
>>>> for a class I'm teaching.  While I'm sure the MITgcm is overkill  
>>>> for this, I'm at least vaguely familiar with it.
>>>>
>>>> I have the winds blowing, the channel seems to behave, and an  
>>>> Ekman layer seems to form.  However, the surface pressure is not  
>>>> behaving and I'm finding the docs on the surface pressure solver  
>>>> opaque enough that I thought I'd ask for help here.
>>>>
>>>> My vertical resolution near the surface is 0.75m, for 40 bins.   
>>>> The total water depth is as deep as 200 m, with sloping sides in  
>>>> a 20 km-wide channel.  dx = 200 m.
>>>>
>>>> What happens is I blow a south wind, the Ekman layers pile water  
>>>> up in the west and move it away from the East. However, the  
>>>> depth anomalies do not seem to get large enough, and the ones  
>>>> that do form
>>>> do not propagate more than a couple of grid cells away at  
>>>> anything like the surface wave speed.  For instance, the cross- 
>>>> channel Ekman flux reaches 1 m^2/s, which should be changing the  
>>>> surface tilt by something like 36 cm over an hour. However,  
>>>> after 7 h or so (in sim time), I still have 10 cm-high anomalies  
>>>> in the grids adjacent to the edge of the channel, and nothing  
>>>> more than 3 grid cells away.
>>>>
>>>> Any hints what I've done wrong or suggestions for getting the  
>>>> barotropic pressure gradient to set up properly across the  
>>>> channel would be very helpful!
>>>>
>>>> Thanks a lot,  Jody
>>>>
>>>> In "data" I hope the relevant parameters are:
>>>>
>>>> viscAz=1.E-3,
>>>> viscAh=1.E-3,
>>>> no_slip_sides=.FALSE.,
>>>> no_slip_bottom=.TRUE.,
>>>> viscA4=0.E12,
>>>>
>>>> gBaro=9.81,
>>>> rigidLid=.FALSE.,
>>>> implicitFreeSurface=.TRUE.,
>>>>
>>>> nonHydrostatic=.FALSE.,
>>>> hFacMin=0.5,
>>>> implicSurfPress=0.5,
>>>> implicDiv2DFlow=0.5,
>>>> nonlinFreeSurf=3,
>>>> hFacInf=0.1,
>>>> hFacSup=2.0,
>>>> exactConserv=.TRUE.,
>>>>
>>>>
>>>>
>>>>
>>>> -- 
>>>> Jody Klymak
>>>> http://web.uvic.ca/~jklymak/
>>>>
>>>>
>>>> _______________________________________________
>>>> MITgcm-support mailing list
>>>> MITgcm-support at mitgcm.org
>>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>>
>>>
>>> -- 
>>> ------------------------------------------------
>>> Dmitri A. Leonov
>>> Research Associate, Postdoctoral
>>> University of Washington, School of Oceanography
>>>
>>>
>>> _______________________________________________
>>> MITgcm-support mailing list
>>> MITgcm-support at mitgcm.org
>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>
>> --
>> Jody Klymak
>> http://web.uvic.ca/~jklymak/
>>
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support