Hi, Hangsheng
You may want to try r.surf.contour first to see if it gives satisfactory
results (use v.to.rast to convert the vector contour lines to raster). It
uses a flood fill approach that is particularly suited to contour line
data. It is much simpler and faster that s.surf.tps, but doesn't allow
the amount of user tuning that s.surf.tps does. Call me if you need any
more details.
-Malcolm Williamson
On Sat, 13 Jul 1996, Hangsheng Lin wrote:
Dear GRASS users:
Is v.surf.tps available now? What is the general accuracy of s.surf.tps?
80%, 90%, or 50% (or whatever number) accurate? If v.surf.tps not ready
yet, to use s.surf.tps to interpolate a contour map to a full surface,
which of the following two approaches yields better accuracy? 1) use
v.to.rast to convert the vector file to a raster map -> use r.random to
sample, say 50% of the points -> use s.surf.tps to interpolate; 2) use
v.to.sites to extract points from the vector file -> use s.surf.tps to
interpolate. If I wish to assess the accuracy of the interpolation, how
should I do it? I am talking about a large area (~1/3 of Arkansas size and
a county size), no field data available for checking in my own study.Any feedback will be greatly appreciated. Thanks in advance.
Sincerely,
Hangsheng Lin