[MITgcm-support] filter for cubesphere-grid-corners

[Andrea Molod]

hi yuan,

good news that changing the numerics for the pressure gradient
term helps with the wave 4 stuff. jean michel will be back next
week and can help you with some info about how to generate the
'better' corner point locations at the resolution you need - i would
guess that this will get rid of most, if not all, of the structures
you are talking about. about ABSOLUTE zonal symmetry, i am not
sure how to get that with this grid....
good luck!
andrea

----------------------------------------------------------------
Andrea M. Molod, PhD.
Department of Earth, Atmospheric & Planetary Sciences
MIT
Tel: (617) 253-5458, Email: amolod at mit.edu
WWW:http://paoc.mit.edu/paoc/people/person.asp?position=Postdoc&who=molod
----------------------------------------------------------------


On Mon, 25 Sep 2006, Yuan Lian wrote:

> Hi, Andrea,
>
> I tested the cube sphere grid with three new options you suggested:
> useAbsVorticity=.TRUE.,
> SadournyCoriolis=.TRUE.,
> selectKEscheme=3,
>
> The atmospheric model works much better now. There are still corner effects 
> (wave number 4 structures) at the beginnning of the simulation but they will 
> be gone fairly soon. The corner effects in my simulation now perform as 
> initial perturbations which can break zonal symmetry.
>
> Thanks!
>
> Yuan
> Yuan Lian wrote:
>
>> Hi, Andrea,
>> 
>> I just compared the tile files. Those that are used in fizhi-cs-aqualev20 
>> are different from what I generated from matlab codes. Could you please 
>> tell me how to generate the new grids with location of corners changed? Or 
>> could you please generate C64 and C128 tile files since I need some high 
>> resoultion grids?
>> 
>> Thanks a lot!
>> 
>> Yuan Lian wrote:
>> 
>>> Hi, Andrea,
>>> 
>>> Thanks for the tips. I am still using old tiles which were generated by 
>>> the matlab codes. I will try these configurations and report the results 
>>> here.
>>> 
>>> Yuan
>>> 
>>> Andrea Molod wrote:
>>> 
>>>> hi yuan,
>>>> 
>>>> this issue at the corners was addressed by jean-michel by changing
>>>> the locations of some of the corner grid points and by changing
>>>> some options in the numerics. it only showed up as something we could see 
>>>> in atmospheric simulations with no stationary waves to speak of. his 
>>>> change made the 'corner' issue go away for us.
>>>> 
>>>> i don't know if you have the 'new' tile files or not, but you can
>>>> find them for sure in the fizhi aqualev20 verification experiment.
>>>> just check to see if they match what you are using, if not, grab
>>>> them. also take a look in the 'data' file for this experiment, you
>>>> will see:
>>>>  useAbsVorticity=.TRUE.,
>>>>  SadournyCoriolis=.TRUE.,
>>>>  selectKEscheme=3,
>>>> hope this makes the need for a 'corner filter' go away....
>>>> good luck.
>>>> andrea
>>>> 
>>>> ----------------------------------------------------------------
>>>> Andrea M. Molod, PhD.
>>>> Department of Earth, Atmospheric & Planetary Sciences
>>>> MIT
>>>> Tel: (617) 253-5458, Email: amolod at mit.edu
>>>> WWW:http://paoc.mit.edu/paoc/people/person.asp?position=Postdoc&who=molod 
>>>> ----------------------------------------------------------------
>>>> 
>>>> 
>>>> On Fri, 22 Sep 2006, Yuan Lian wrote:
>>>> 
>>>>> Hi, Ed
>>>>> 
>>>>> I am trying to add an extra corner filter (like zonal filter as a polar 
>>>>> filter) to the eight cubesphere-grid corners. I am thinking to add this 
>>>>> filter as a subroutine so its strength is adjustable. The reason why I 
>>>>> will do this is because I found the corner effect is very obvious in my 
>>>>> atmospheric model. Could you please give me some instructions to do 
>>>>> this? Thanks a lot!
>>>>> 
>>>>> Yuan
>>>>> _______________________________________________
>>>>> MITgcm-support mailing list
>>>>> MITgcm-support at mitgcm.org
>>>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>>>> 
>>>> _______________________________________________
>>>> MITgcm-support mailing list
>>>> MITgcm-support at mitgcm.org
>>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>>> 
>>> 
>>> _______________________________________________
>>> MITgcm-support mailing list
>>> MITgcm-support at mitgcm.org
>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>> 
>> 
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>> 
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>