SBL wrote > Hei Helumt, > > In that case I would probably try r.param.scale in order to identify flat > areas at maybe different scale levels. > > V-valleys have a higher variation in aspect at the bottom (sides are > almost facing each other and nearly "meet" at the lowest point in a > V-valley). While sides are more or less steep. So slope and aspect at a > coarser level (also available through r.param.scale) could help to > distinguish V- and U-valleys. > > Depending on scale U-Valleys have a TPI of ~0 at the bottom and ~0 at the > middle of the sides (equaly much terrain above and below a focal cell in > the given neighborhood). > > Some background material on TPI can be found here: > http://www.jennessent.com/arcview/tpi.htm > Jenness used ArcView. Yet TPI is a simple combination of r.neighbors and a > mapcalc. > > Good luck with your modelling.
as a possible additional parameter for such a modelling, I've tried to implement a Multi-resolution Valley Bottom Flatness (MrVBF) index. http://lists.osgeo.org/pipermail/grass-commit/2014-June/031834.html Author: hellik Date: 2014-06-22 16:04:15 -0700 (Sun, 22 Jun 2014) New Revision: 60919 Added: grass-addons/grass7/raster/r.valley.bottom/ grass-addons/grass7/raster/r.valley.bottom/Makefile grass-addons/grass7/raster/r.valley.bottom/r.valley.bottom.html grass-addons/grass7/raster/r.valley.bottom/r.valley.bottom.py Log: r.valley.bottom: a script to calculate a Multi-resolution Valley Bottom Flatness (MrVBF) index testing and hints for improvement are very welcome. thanks ----- best regards Helmut -- View this message in context: http://osgeo-org.1560.x6.nabble.com/calculation-valley-floor-width-in-GRASS-or-some-index-indicating-valley-floor-width-tp5132402p5147233.html Sent from the Grass - Users mailing list archive at Nabble.com. _______________________________________________ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user