More help on v.circle, please

[ish at alla.pl.swf.usace.army.mil]

>2.  When v.support is run the resultant circle vector file, the labels
>     that we generate are lost for overlapping polygons.  We don't
>     want this.
>
>3.  What we would eventually like is information on any point in the
>     area as to how many circle (or in this case, 4km buffers around
>     grouse habitat sites) intersect the point.  We want to know how
>     many buffers a point would have intersecting it.
>

David,

I wrestled with the overlapping polygon dilemma for a similar GIS
application involving overlapping sitings of threatened and
endangered species from the Nature Conservancy's Biological Conservation
Database (BCD). There may be a more eloquent way to do this, but here's
what worked for us:

1 - subset your vector data into as many maplayers necessary to 
    achieve nonoverlap so that you can successfully run v.support,
2 - rasterize each of the vector maplayers (v.to.rast),
3 - run r.cross on the set of rasterized polygons which creates a new
    raster map where each overlap is now a unique category with a label
    describing the multiple memberships derived from the input maplayers.
4 - if you want vectors, run GRASS command r.poly to generate a vector
    version of the new raster file (you may need to copy the cats file from
    the raster map to the dig_cats directory).

In our application, we linked this vector file to our T&E database and
used dbtools (GRASS contributed code from CAST) to execute SQL programs
to do queries to report on and redraw the vectors based on the categories.
Contact me directly if you would like additional info. Good luck.

Ishmael Williams
US Army Corps of Engineers - Fort Worth District
(817) 885-6391
email: ish at alla.pl.swf.usace.army.mil