On Thu, Nov 15, 2012 at 6:00 PM, Moritz Lennert <[email protected]> wrote: > On 15/11/12 14:08, Newcomb, Doug wrote: >> >> I guess the question is, if you then performed v.rast.stats for an >> underlying raster using the overlapping buffers generated with -t , how >> would the statistics for the overlapped polygons work? Think 2 1km >> circles whose centers are .75km apart. > > > The -t flag (thanks for that, btw !) allows to not fusion the buffers, but > it still respects topology, i.e. no overlaps allowed in polygons. The parts > where buffers overlap are separate polygons from the rest of the buffer, and > they have more than one category (similar to the result of importing a > shapefile with overlapping polygons).
The problem is that v.rast.stats does not support areas with multiple categories in the same layer, because v.to.rast does not support that. In this case, buffers should be created for each feature separately using the cats or where options. Markus M _______________________________________________ grass-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-dev
