Re: Grid interpolation

[Billy Kessler <William.S.Kessler@xxxxxxxx>]

The problem here is usually in trying to compute fluxes, budgets and the like from model output. You 
would like to have the results be a decent representation of what the model is actually doing. In my 
experience, getting these things right requires computing the derivatives EXACTLY the way the model 
did, which often involves high-order derivative schemes. If you don't do that then you can get the 
wrong answer (for example in an estimate of advective fluxes), as I have found out from unfortunate 
experience. So the question is deeper than simply how to put the variables on the same grid.

As simple first-guess solution is to define a grid twice as dense as the originals, so every u,v,m 
point lies on the new grid. Then interpolate using @fln. That'll usually get you pretty close to the 
answer, but you really should check the model documentation about how it takes derivatives, 
otherwise don't be surprised if budgets don't balance very well.

Billy K


> Date: Tue, 26 Apr 2005 12:25:42 -0500 (CDT)
> From: jimmyc@iastate.edu
> To: ferret_users@noaa.gov
> Subject: Grid interpolation
> 
> Hello-
> I have a staggered grid where the mass, u and v wind components are all on 
> different axes.
> 
>   M U M U M
>   V   V   V
>   M U M U M
> so that my axis for the M grid is i=1:350 j=1:308
>                          U         i=1:350 j=1:307
>                          V         i=1:349 j=1:308
> Fun stuff right?
> How does one attempt to interpolate each wind component to the M grid? I 
> am not worried about the end points as I plan to avoid using them.
> Any insight is greatly appreciated.
> 
> 
> 
> -- 
> James Correia Jr.
> Ph.D. Candidate in Agricultural Meteorology
> President (Dictator) of the ISU Graduate Meteorology Club
> Dept. of Agronomy, Iowa State University
> *********************************************
> Email: jimmyc@iastate.edu
> Phone: 515-294-5587
> Web: http://www.mesoscale.iastate.edu/jimmyc/
> ISU GMC: http://www.stuorg.iastate.edu/gmc/
> *********************************************
> "There is a difference between knowing the path and walking the path"
> -"Morpheus"
> 
> "Scientific theories are built to be used, but they must be mistrusted, 
> tested and improved."
> -Balik and Frank