Re: [ferret_users] [Ferret_Users] SAMPLEIJ of an array of size[680,380]

[Jaison Kurian <jaison@xxxxxxxxxxxxxxxxxx>]

Hi Zhen,
          Yes, SAMPLEIJ needs same length for the ipts and jpts. The
resulting variable from this external function will have only one
abstract axis which corresponds to the length of ipts(=jpts) instead
of X and Y (Z and T will be inherited from the original variable).

  From the values given for ipts (=2,4,6,....,680) & jpts (=2,4,6,..
..,380) it seems that you are trying to regrid the density variable
to a coarser grid (without any averaging) which does have alternate
positions as grid points. For doing this there are proper methods like
--> defining a coarser destination grid, regrid the variable onto
that grid using @XACT regridding. If you can explain what exactly
you want to do, I can offer you a better help.

Let me know if you have any questions.

Regards,

Jaison

On Wed, 31 May 2006 zhen.li@gsfc.nasa.gov wrote:

> Dear Ferret_Users:
>
> It seems to me that SAMPLEIJ only subsamples the array that has the same 
> number of
> points in "i" and "j" direction. For example, my effort to use the 
> following
> script to subsample the array of size [680,380] failed, because it only 
> subsampled
> the point in the "I" direction. The new array "var_ij" has a size of 
> [390,1]. I wonder
> if anyone can help me with this case.
>
> !FILE/VAR="ipts" ipts.dat (ipts=2,4,6,....,680)
> !FILE/VAR="jpts" jpts.dat (jpts=2,4,6,....,380)
> !use ocean_density.res.nc
> !let var_ij=sampleij(RHO_TAUM1,ipts[d=1],jpts[d=2])
>
> Thanks!
>
> Zhen