addons/r.convergence

[selvaje]

Hi I noticed a strange behavior of r.convergence in some tiles of my digital elevation model. In few words if I process a single tile I obtain a convergence result if I combine 4 tiles and rerun r.convergence I obtain another result. Such behavior appear only in few tiles along the same longitudinal gradient (see pdf). Seams to me that is something related with the bounding box of the tiles and the calculation of the distance. I'm doubting that is something related to the tiling system because I have computed several other topographic variables and the problem appear only with r.convergence.

I attach the tar.gz with the data and the script to reproduce the error.

Thanks for the support Giuseppe GRASS GIS 7.3.svn

[selvaje]

files wget ​https://www.dropbox.com/s/72ndnxtpiqqhmvx/tmp.tar.gz wget ​https://www.dropbox.com/s/aw5szez5xspqjon/convergence_10KMmedian_MERIT.pdf

[martinl]

Remove Milestone from Addons bugreports.