Re: grid rotation

Hello

You are correct about unbounded regions.

http://www.ncl.ucar.edu/Applications/grid_fill.shtml
In particular, Examples 3 & 4

It may be possible to do something like:

[1] http://www.ncl.ucar.edu/Applications/lat2pop.shtml
     Interpolate to a (say) POP grid.

[2] Use poisson_grid_fill

[3] Interpolate back to the 1x1 grid

[4] Force all the original non-missing values onto
     the interpolated grid [3]

===

I have not seen the distribution of the 1x1 missing values.

Possibly create a larger grid with 'extra' pole grid points.
These could be set to some avarage.

Good luck

Jan Sedlacek wrote:
> Hi
>
> I have a global 2-D 1deg_x_1deg scalar field with missing data and I
> would like to fill the gaps. I discovered the function poisson_grid_fill
> and would like to use it. However, I'm mostly interested in the Arctic
> region and since at the Pole the boundaries are closed it does not use
> data from across the Pole to interpolate. I thought that maybe I could
> rotate the whole grid, so that the North Pole would be somewhere near
> the Equator.
>
> I found in the archives several discussions about changing the Pole
> location for plotting (i.e., CenterLonF,..) but not for calculating. Is
> there a simple way to do it without having to code the whole coordinate
> transformation?
>
> Thanks a lot,
> Jan
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Feb 16 16:54:37 2010