Hi
On 12/13/2013 01:14 PM, image wrote: Hello, Thank you very much for your reply. I made some tests following your recommandations. For information, the surface of my raw raster DEM is 15.000 km². Resolution : 1 pixel X & pixel Y = 30 meters. (SRID 2154 UTM).1/"r.watershed": the main input parameter seems to be the minimum size for each basin. I typed 10 (unit pixels i think). I think it's a low value. I would like know how enter a coherent value. How to ensure that our input value is coherent? You choose the threshold value as the minimum size of drainage area that is of interest. In your case, with resolution 30m, then 1 pixel = 900 sq.m. So a threshold of 11000, for example, would give a minimum drainage area of 10 sq km. (A threshold of 10 is way too small!) The names of the basin and stream rasters are whatever you entered as the "basin=..." and "stream=..." parameters2/"r.watershed" process generates several new raster layers. Could you confirm to me that for *streams* r.thin & r.to.vect process i have to select the third layer generated called "water_segments". And for the basins i have to select the 4th layer called "Basin label"? I suggest that you have a look at the manual page: http://grass.osgeo.org/grass65/manuals/r.watershed.html Perhaps for this job you could try the actual GRASS interface. It's a bit harder to learn, but well worth the effort...3/For basins and the v.extract process. I made a test with v.extractwhere. But i don't find how to enter the value type=boundary. For information i'm working with easy GUI interface via Qgis., no with command lines interface. Change the threshold value... If you need a very detailed stream layer, you can run the module once with a small thresh for streams, then a second time with a large thresh to delineate the basins.4/ The final vector output data for streams seems to be rather coherent. However, my basins vector segmentation seems to be really too detailed and inexploitable. And one more thought: If you can manipulate the original DEM, so that areas with zero value or no values (all the lower left corner, etc), are actually set to NULL, then you will avoid those funny straight stream segments. Please find below usefull SS showing : A: My raster raw data / B:Streams_vect_3D / C: Basins_labels_raster D: Basins_area_vector A <http://osgeo-org.1560.x6.nabble.com/file/n5094142/SS_raw_DEM.jpg> B <http://osgeo-org.1560.x6.nabble.com/file/n5094142/SS_streams_vect_3D.jpg> C <http://osgeo-org.1560.x6.nabble.com/file/n5094142/Basins_label_raster.jpg> D <http://osgeo-org.1560.x6.nabble.com/file/n5094142/Basin_area_vect.jpg> If you can throw light for me. Thanks. With kind regards. -- View this message in context: http://osgeo-org.1560.x6.nabble.com/Extraction-3D-vectors-ridges-Thalweg-from-raster-DEM-tp5093957p5094142.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 This mail was received via Mail-SeCure System. -- Micha Silver GIS Consultant, Arava Development Co. http://www.surfaces.co.il |
_______________________________________________ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user