Maciej Sieczka wrote:
Luigi Ponti wrote:
What I meant is that the outer Voronoi cells may extend up to the
region's bounding box, with consequent new Voronoi cells for
interpolation points outside the convex hull, please see
http://quartese.googlepages.com/voronoi
Anyway - what could be done to overcome the trimming to convex hull
effect: you need to make sure that the convex hull for your input cells
covers your whole area of interest - ie. enlarge region enough for
important input cells not to be omitted. If cells in the input raster
are distributed so that a convex hull covering the whole area of
interest is not possible, you need to enhance your input raster.
Please, have also a look at
http://www.ems-i.com/gmshelp/Interpolation/Interpolation_Schemes/Natural_Neighbor_Interpolation.htm
This web page suggests an approach for extrapolation that would get a
non-null value to all the points in the region: four additional input
points are included in the NN interpolation that correspond to the
corners of a user-defined region (a bounding box that includes all input
points), by assigning them a value based on a different interpolation
method (IDW). This way, the convex hull of input points has the same
extent of the bounding box of the region. Also, I think that there is no
need of giving the -W nnbathy flag a negative value and the resulting
interpolated raster would still be inside the range of the input data
because of the way IDW works.
It sounds feasible to add this functionality to r.surf.nnbathy, but I
don't know if the approach is correct from the GIS point of view (i.e.,
mixing two interpolation methods) or even more useful/better than just
using IDW for the whole region.
Thanks and regards,
Luigi
_______________________________________________
grassuser mailing list
[email protected]
http://grass.itc.it/mailman/listinfo/grassuser
