[GRASS-user] v.clean

Roger Mason rmason at esd.mun.ca
Tue Aug 5 06:50:30 EDT 2008


I'm still working on generating fake geological maps and have made
significant progress.  I generate a contour map of a fake
topography made using r.surf.fractal and superimpose geological
boundaries made using r.plane.  Both these steps work well.  I need to
find intersections of topographic contours and contours of the
geological boundaries.  I should be able then to select one
intersection for each geological boundary and place a marker on the
output map (made using GMT).  I am using v.patch
to combine the contours of geological boundaries with the topographic
contours and then using v.clean to find intersections of features in
the combined vector map.  The code is below.

    r.plane name=plane${j} dip=${dip} azimuth=${azimuth} \
      easting=${kmax} northing=${lmax} elevation=${hmax} \
      type=double --o --q
# This necessary to find intersections of landscape and
# geol. boundaries.
    r.contour input=plane${j} output=plane${j}contours \
      minlevel=${min} maxlevel=${max} step=${step} --o --q
    v.patch input=smoothedcontoured,plane${j}contours \
      output=plane${j}topo --o --q
# I got the next line from the v.patch docs.  It is'nt reliable, it
# often finds spurious intersections.
    v.clean input=plane${j}topo output=plane${j}intersections \
      error=plane${j}err tool=snap,break,rmdupl --o --q
# Write a GMT file with interesections of land surface and
# geol. boundaries.  The files are called plane1err.gmt etc
    v.out.ogr input=plane${j}err type=point format=GMT \

The problem is that v.clean often finds spurious intersections,
i.e. points that don't correspond to intersections of contours of
geological surfaces and the topographic contours.

If someone can see what I'm doing wrong or knows of a more reliable
method to accomplish my goal please tell me.  If you need more
information please ask.


Roger Mason
Grass 6.3.0
Gentoo Linux

More information about the grass-user mailing list