[MITgcm-support] RE: cosPower

[Chris Hill]

Hi Ichiro,

 Good to hear from you. I saw the quarter degree movie being projected large
at several booths in Phoenix - it looked great! 

 Analytically it looks like it might want to be four for biharmonic (i.e.
the effective damping time scales as the fourth power of the resolution).
However because of issues like the net damping being the sum of two
directions and bleed over several time steps from neighboring points where
resolution is maybe higher or lower the analytic number is usually reduced
slightly, typically 3 - 3.5. 

 For an isotropic grid you need to set the USE_ISOTROPIC_COS_SCALING flag in
GAD_OPTIONS.h. 

 Looking at the code it looks like in a mixed del^2, del^4 scenario you
would need to create a couple of extra cosFacU4 and cosFacV4 arrays, set
them in ini_spherical_polar_grid.F and use them in the four places where
viscA4 is used. I could be wrong on this, there could be a trick that lets
you do it already - I'm cc'ing this to support to see if anyone has another
answer.

Chris

> -----Original Message-----
> From: Ichiro Fukumori [mailto:fukumori at jpl.nasa.gov] 
> Sent: Wednesday, November 19, 2003 8:43 PM
> To: cnh at mit.edu; adcroft at ocean.mit.edu
> Cc: Dimitris Menemenlis; tlee at pacific.jpl.nasa.gov; Detlef Stammer
> Subject: cosPower
> 
> Chris and Alistair,
> 
> Our 1/4-deg model has grid-scale noise that we suspect is due 
> to inadequate viscosity.
> We are using biharmonic friction with viscA4=1.e10.  There is 
> a cosPower parameter
> set in the data file.   We are using a value of 3.  We think 
> cosPower is 
> adjusting the
> magnitude of friction parameter to different grid size.  If 
> that is so, I think this value should
> be 4 for biharmonic and 2 if Laplacian were used.   What 
> should cosPower be?
> What if we want to somehow use both biharmonic and Laplacian?
> 
> Thanks, Ichiro
> 
> 
> ===============================================================
> Ichiro Fukumori               | Jet Propulsion Laboratory
> e-mail: fukumori at jpl.nasa.gov | Mail Stop 300-323
> phone: +1 (818) 354-6965      | 4800 Oak Grove Drive
> fax: +1 (818) 393-6720        | Pasadena, California 91109, U.S.A.
> ===============================================================
>