Date: Thu, 16 Jul 92 13:31:59 EST
From: simon@cerberus.earth.monash.edu.au (Simon Cox)
Message-Id: <9207160331.AA02871@cerberus.earth.monash.edu.au>
Sender: lists-owner@amber.cecer.army.mil
Reply-To: grassu-list@amber.cecer.army.mil
Precedence: Bulk
To: grassu-list@amber.cecer.army.mil
Subject: interpolation programsAre there any interpolation programs for grass out there EXCEPT
the idw pair?Inverse distance weighting may be fast and easy, but
it does not produce very good results for some types of problem -eg if you have a "surface" which you *know* is smoothly varying,
(eg geophysical data)
then some form of surface fitting (polynomial, spline, etc) can be
better - it allows you to find peaks between the data points, whereas the
idw algorithm will NEVER produce a value higher (or lower) than
the surrounding values.Thanks for your assistance
Simon Cox
Dept of Earth Sciences
Monash University
Clayton, Vic 3168 Australia
We have a program which uses thin plate splines. I have been meaning to
put it out for distribution but everytime I think I will, the programmers
come up with another major enhancement. I will try to get it out
this week for you. It is called s.surf.tps. It uses site
file data. I will include the source for a program which will dump
out every point from a labeled vector file as site info. With that, you
can convert contour data to site to be used as input.
--
Dave Gerdes
US Army Construction Engineering Research Lab
Spatial Analysis & Systems Team
dpgerdes@cerl.cecer.army.mil
(217) 352-6511 x591