Re: gridding non gridded data

[Med Bennett <mbennett(at)indra.com>]

Teresa Sanchis wrote:

> Dear Colleagues,
>
> I'm trying to get a 3-D regular grid from my non gridded data.
> First I tried with GRID3, but my results made no sense because I got
> values much bigger or smaller than my non gridded values.
> I have also been surfing in internet and I found a routine which uses
> TSC (Triangular Shaped Cloud interpolation of irregularly gridded data
> onto a regular grid). The problem is that I tested this routine and found
> that it's necessary a great number of points to get a good gridding.
> Unfortunately I only have few points (about a hundred) with many "holes"
> between them.
>
> I'm writing to you because maybe you know another type of
> gridding routine that could help me.
>
> Thanks in advance
>
> Teresa Sanchis
>

Hi Teresa,

You can write your own 3D inverse distance squared interpolator, or look into
a variety of geostatistical packages that do gridding and interpolation in
three dimensions.  There are many canned programs that do such things but
they are usually expensive.  You can track down GSLIB freeware FORTRAN code
on the net - it might not be too hard to translate into IDL.  I just use the
fortran code and use IDL for I/O and visualizing results.  Good luck!