[Numpy-discussion] Vector interpolation on a 2D grid (with realistic results)

Ryan May rmay31 at gmail.com
Sat Nov 7 21:11:00 EST 2009

On Sat, Nov 7, 2009 at 5:38 PM, Pierre GM <pgmdevlist at gmail.com> wrote:
> Linear interpolation with the delaunay package doesn't work great for
> my data. I played with the radial basis functions, but I'm afraid
> they're leading me down the dark, dark path of parameter fiddling. In
> particular, I'm not sure how to prevent my interpolated values to be
> bounded by the min and max of my actual observations.
> Ralf' suggestion of smoothing the values afterwards is tempting, but
> sounds a bit too ad-hoc (btw, Ralf, those were relative differences of
> monthly average precipitation between El Niño and Neutral phases for
> November).

That was me, not Ralf. :)  And I agree, I the interpolated field does
look a bit noisy for such data.  I've been doing the smoothing on top
of natural neighbor for doing some of my own meteorological analysis.
Using the Gaussian kernel isn't really *that* ad hoc considering the
prevalence of Barnes/Cressman weighting for spatial averaging
typically used in meteorology.  And if you have no idea what I'm
talking about, Google them, and you'll see. :)


Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

More information about the NumPy-Discussion mailing list