thinning algorithms in GRASS

[Gillian Bowser]

Hi all!

I am curious about the accuracy/precision of translating raster files to vector
files. I am creating paths from point A to B and I need to know the length of
the path followed between those points.  The paths are being generated by
building first a cost surface, then draining (r.drain) that surface.  To
run regression on the paths, I translated r.drain through r.thin to a 
vector file which was exported to Splus.

Whew!  The problem is I have a huge variance around path lengths and am trying
to explain the source of that variance--is it due to the thinning process,
the original cost construction, etc.?  Does anyone know how precise the 
thinning algorithms are and how do you translate any error into a number
(is it additive with the RMS of the underlying surface???)


Any insights would be greatly appreciated!


Gillian Bowser
gillian at rmro.nps.gov