Re: [GRASS-user] elevation profile
Thanks to everyone for their responses. I did not try all the methods. One method that did work (and was mentioned more than once) was basically: v.to.points (convert 2d lines to points) v.out.ascii (export coordinates to txt file) cat "coords.txt" | r.profile (input text file coords to r.profile) However, this worked for only one line/path at a time (by using SQL where clause with an attribute unique to each path) and I had a shapefile with about 20 paths. Likely this could have been scripted to process efficiently. Unfortunately, I'm still learning the basics of GRASS and did not want to get into scripting yet! I went with SAGA in the end. Thanks Ricardo for that suggestion! It worked great and had exactly what I wanted. Also, it's a good idea to start with a clean line vector dataset, regardless of which method you choose. It took a fair bit of time to snap/merge several line segments together to get my paths to come out right. Thanks again. - John ** John Callahan, Geographic Information Scientist Delaware Geological Survey, University of Delaware URL: http://www.dgs.udel.edu ** On Fri, Jul 16, 2010 at 4:15 AM, Francesco Mirabella wrote: > Hi, > I had a similar issue while extracting an elevation profile for a river. I > think it's similar to your problem. > > Afaik v.drape produces x,y,z vector map, if you need to export it, maybe > you should use v.out.ascii. However, v.drape gives coordinates of points, > NOT the distance between each point which I think you need if you want to > get an elevation profile (distance vs elevation). To do this I suggest the > following: > > v.out.ascii of your 2d vector (road) > cat output_of_v.out.ascii| r.profile input=your_dem > output=elevation_profile > > Hope it helps > cheers > Francesco > > > > > > Hamish wrote: > >> yet another two methods to try- >> >> (spearfish dataset) >> >> # v.out.ascii | r.profile method >> >> g.region -d >> v.in.region sp_box >> v.type in=t_powerlines out=t_powerline type=boundary,line >> v.overlay ain=t_powerline bin=sp_box out=sp_powerline atype=line op=and >> v.build.polylines in=sp_powerline out=sp_powerline_pl >> g.remove v=sp_box,t_powerline,sp_powerline >> >> v.out.ascii in=sp_powerline_pl format=standard | grep '^ ' | \ >> r.profile -g in=elevation.dem res=30 >> >> >> >> # v.to.rast MASK method >> >> v.type in=t_powerlines out=t_powerline type=boundary,line >> v.category in=t_powerline out=t_powerline_with_cat option=add >> v.to.rast in=t_powerline_with_cat out=MASK use=val value=1 >> g.remove v=t_powerline,t_powerline_with_cat >> >> r.out.xyz elevation.dem >> g.remove MASK >> >> >> >> ___ >> grass-user mailing list >> grass-user@lists.osgeo.org >> http://lists.osgeo.org/mailman/listinfo/grass-user >> > > > > ___ > grass-user mailing list > grass-user@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/grass-user > ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] elevation profile
Hi, I had a similar issue while extracting an elevation profile for a river. I think it's similar to your problem. Afaik v.drape produces x,y,z vector map, if you need to export it, maybe you should use v.out.ascii. However, v.drape gives coordinates of points, NOT the distance between each point which I think you need if you want to get an elevation profile (distance vs elevation). To do this I suggest the following: v.out.ascii of your 2d vector (road) cat output_of_v.out.ascii| r.profile input=your_dem output=elevation_profile Hope it helps cheers Francesco Hamish wrote: yet another two methods to try- (spearfish dataset) # v.out.ascii | r.profile method g.region -d v.in.region sp_box v.type in=t_powerlines out=t_powerline type=boundary,line v.overlay ain=t_powerline bin=sp_box out=sp_powerline atype=line op=and v.build.polylines in=sp_powerline out=sp_powerline_pl g.remove v=sp_box,t_powerline,sp_powerline v.out.ascii in=sp_powerline_pl format=standard | grep '^ ' | \ r.profile -g in=elevation.dem res=30 # v.to.rast MASK method v.type in=t_powerlines out=t_powerline type=boundary,line v.category in=t_powerline out=t_powerline_with_cat option=add v.to.rast in=t_powerline_with_cat out=MASK use=val value=1 g.remove v=t_powerline,t_powerline_with_cat r.out.xyz elevation.dem g.remove MASK ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] elevation profile
yet another two methods to try- (spearfish dataset) # v.out.ascii | r.profile method g.region -d v.in.region sp_box v.type in=t_powerlines out=t_powerline type=boundary,line v.overlay ain=t_powerline bin=sp_box out=sp_powerline atype=line op=and v.build.polylines in=sp_powerline out=sp_powerline_pl g.remove v=sp_box,t_powerline,sp_powerline v.out.ascii in=sp_powerline_pl format=standard | grep '^ ' | \ r.profile -g in=elevation.dem res=30 # v.to.rast MASK method v.type in=t_powerlines out=t_powerline type=boundary,line v.category in=t_powerline out=t_powerline_with_cat option=add v.to.rast in=t_powerline_with_cat out=MASK use=val value=1 g.remove v=t_powerline,t_powerline_with_cat r.out.xyz elevation.dem g.remove MASK ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] elevation profile
This might be useful: http://casoilresource.lawr.ucdavis.edu/drupal/node/375 Cheers, Dylan On Thursday 15 July 2010, John Callahan wrote: > Hopefully a quick question that I'm struggling with. (I'm a long-time GIS > user but trying to move my work to GRASS and other FOSS4G.) I'm using > GRASS 6.4 RC6 on Windows XP. > > I have a raster grid representing elevation at 2meter resolution. I have a > line shapefile representing roads (about 20 roads total). I would like to > get ASCII text files of x,y,z at about 1 or 2 meter intervals along each > road. Any ides of what I can try? > > I thought v.drape might work. This converts my 2D road shapefile to 3D, but > I'm not sure where to go from there. And this doesn't help getting 1 or 2 > meter intervals. I also tried r.profile but this requires input as a > listing of x,y coordinates, not shapefiles. I know there's a step here I'm > missing. > > - John > > ** > John Callahan, Geographic Information Scientist > Delaware Geological Survey, University of Delaware > URL: http://www.dgs.udel.edu > ** -- Dylan Beaudette Soil Resource Laboratory http://casoilresource.lawr.ucdavis.edu/ University of California at Davis 530.754.7341 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] elevation profile
Hi I think r.profile[1] can probably help you get to the output you want. In the past I had looked into doing the exact same thing as you. I ended up using SAGA GIS[2], as it has a predefined module that does exactly this task, and I was in a hurry ;) Good luck [1] - http://grass.itc.it/grass64/manuals/html64_user/r.profile.html [2] - http://www.saga-gis.org On Thu, Jul 15, 2010 at 8:41 PM, John C. Tull wrote: > Hi John, > You will likely get a more complete answer from others, but you seem to be > on the right track with v.drape. From there, you can add a column for > cumulative length with v.db.addcol, then run v.to.db with option=length to > add a length column to your 3d vector. You might be able to subsample from > there. > Another possibility would be to run v.segment to generate points at your > desired length interval for the original vector. Then you could run v.drape > on the resulting points file and work from there or somehow pipe the point > coordinates into r.profile. > Good luck, > John > On Jul 15, 2010, at 12:25 PM, John Callahan wrote: > > Hopefully a quick question that I'm struggling with. (I'm a long-time GIS > user but trying to move my work to GRASS and other FOSS4G.) I'm using > GRASS 6.4 RC6 on Windows XP. > > I have a raster grid representing elevation at 2meter resolution. I have a > line shapefile representing roads (about 20 roads total). I would like to > get ASCII text files of x,y,z at about 1 or 2 meter intervals along each > road. Any ides of what I can try? > > I thought v.drape might work. This converts my 2D road shapefile to 3D, but > I'm not sure where to go from there. And this doesn't help getting 1 or 2 > meter intervals. I also tried r.profile but this requires input as a > listing of x,y coordinates, not shapefiles. I know there's a step here I'm > missing. > > - John > > ** > John Callahan, Geographic Information Scientist > Delaware Geological Survey, University of Delaware > URL: http://www.dgs.udel.edu > ** > ___ > grass-user mailing list > grass-user@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/grass-user > > > ___ > grass-user mailing list > grass-user@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/grass-user > > -- ___ ___ __ Ricardo Garcia Silva ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] elevation profile
Hi John, You will likely get a more complete answer from others, but you seem to be on the right track with v.drape. From there, you can add a column for cumulative length with v.db.addcol, then run v.to.db with option=length to add a length column to your 3d vector. You might be able to subsample from there. Another possibility would be to run v.segment to generate points at your desired length interval for the original vector. Then you could run v.drape on the resulting points file and work from there or somehow pipe the point coordinates into r.profile. Good luck, John On Jul 15, 2010, at 12:25 PM, John Callahan wrote: > Hopefully a quick question that I'm struggling with. (I'm a long-time GIS > user but trying to move my work to GRASS and other FOSS4G.) I'm using GRASS > 6.4 RC6 on Windows XP. > > I have a raster grid representing elevation at 2meter resolution. I have a > line shapefile representing roads (about 20 roads total). I would like to > get ASCII text files of x,y,z at about 1 or 2 meter intervals along each > road. Any ides of what I can try? > > I thought v.drape might work. This converts my 2D road shapefile to 3D, but > I'm not sure where to go from there. And this doesn't help getting 1 or 2 > meter intervals. I also tried r.profile but this requires input as a listing > of x,y coordinates, not shapefiles. I know there's a step here I'm missing. > > - John > > ** > John Callahan, Geographic Information Scientist > Delaware Geological Survey, University of Delaware > URL: http://www.dgs.udel.edu > ** > ___ > grass-user mailing list > grass-user@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/grass-user ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] elevation profile
Hopefully a quick question that I'm struggling with. (I'm a long-time GIS user but trying to move my work to GRASS and other FOSS4G.) I'm using GRASS 6.4 RC6 on Windows XP. I have a raster grid representing elevation at 2meter resolution. I have a line shapefile representing roads (about 20 roads total). I would like to get ASCII text files of x,y,z at about 1 or 2 meter intervals along each road. Any ides of what I can try? I thought v.drape might work. This converts my 2D road shapefile to 3D, but I'm not sure where to go from there. And this doesn't help getting 1 or 2 meter intervals. I also tried r.profile but this requires input as a listing of x,y coordinates, not shapefiles. I know there's a step here I'm missing. - John ** John Callahan, Geographic Information Scientist Delaware Geological Survey, University of Delaware URL: http://www.dgs.udel.edu ** ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from i ntersecting shapefiles
georgew wrote: Much progress! My only remaining question is: Is there a way to use the track raster file to input the values instead of using the mouse, so that the process can be more or less automated? Thank you all for your great help. I'll write a summary and post it in the Wiki as an FAQ or short tutorial as suggested by Markus Neteler. Reading the r.profile man page, you will see that it's possible to pipe the list of coordinates to r.profile: The *profile* parameter can be set to comma separated geographic coordinates for profile line endpoints. The interactive flag (*-i*) overrides this option. Alternatively the coordinate pairs can be piped from stdin, one comma separated pair per line. You can get the coordinate pairs from your track vector with v.out.ascii, and use the fs=, option to make the output file "comma-separated". -- Micha ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from intersecting shapefiles
georgew wrote: Thanks Micha, here we go: Micha Silver wrote: How many shapefiles did you have in the GRASSDATA directory? provide the boundaries. Here is the result: r.info contours_rast ++ | Layer: contours_rast Date: Thu Dec 4 08:50:19 2008 | | Mapset: test02 Login of Creator: george | | Location: test02 | | DataBase: /home/george/GRASSDATA | | Title: Labels ( contours_rast ) | | Timestamp: none | || | | | Type of Map: raster Number of Categories: 0 | | Data Type: DCELL | | Rows: 20 | | Columns: 20 | | Total Cells: 400 | | Projection: x,y | | N: 6020724.5 S: 6013982.586 Res: 337.0957 | | E: 2481099.136 W: 2467034.878 Res: 703.2129 | | Range of data: min = 520.00 max = 1840.00 | | | | Data Source: | | Vector Map: [EMAIL PROTECTED] in mapset test02 | | Original scale from vector map: 1:1 | | | | Data Description: | | generated by v.to.rast | | | | Comments: | | v.to.rast input="[EMAIL PROTECTED]" output="contours_rast" use="attr" \ | | type="line" layer=1 column="ELEVATION" value=1 rows=10 | | | ++ GRASS 6.3.0 (test02):~ > g.region -p projection: 0 (x,y) zone: 0 north: 6020724.5 south: 6013982.586 west: 2467034.878 east: 2481099.136 nsres: 337.0957 ewres: 703.2129 rows: 20 cols: 20 cells: 400 I hope all this helps. Your help so far has been invaluable, thanks a lot. PS I tried to play around with resolution in the Region dialog but the display never changed from the same chunky coloured squares. Here's What I can make out from the above: The X-Y values are about 6,000,000 N and about 2,450,000 E, and the size of your region is about 14,000 X 6500. Let's *assume* these numbers are in meters (UTM projection??). But your resolution is set to 337 e-w and 703 n-s, thus giving you a raster of 20X20 cells. That's certainly too coarse. I'd suggest: Blow away the rasters you made so far. Set resolution to 10m. X 10m. with g.region -s -p res=10 You should then see about 1400 columns by 650 rows =~ 91 cells. Now recreate those rasters. You should get better looking results. -- Micha ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: Re: [GRASS-user] Elevation profile from inter secting shapefiles
Hamish wrote: > A symptom of the computational region being way out of whack compared to > your display region is that you see the result as a series of huge blocks > (computational resolution is too coarse), or in the other direction the > resolution is so fine that takes ages to load. YMMV Note that a mismatch between computational and display regions isn't necessarily "wrong". You may want to perform computations at a resolution finer than that of the display for accuracy, or coarser than it for speed. -- Glynn Clements <[EMAIL PROTECTED]> ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from i ntersecting shapefiles
On 04/12/08 09:47, Hamish wrote: georgew wrote: Hamish, please forgive my ignorance but I am not sure what you mean. As far as I know g.region is the means to change the resolution, correct, whether through the command line or the GUI. Can you please explain how to access and change the computational region settings. I was talking about when using it from the GUI. from the command line the display (d.mon) resolution is the computation region. in the GUI there are two: the active display region(+resolution) and the computational one. What you see in the display window(s) is independent of what's in the $MAPSET/WIND file. That WIND(ow) file holds the region which will be used by any modules running. The reason for all this confusion is that you can have multiple GUI display windows open at the same time, each independently zoomed to a different area and set at a different resolution (for a visual display there's no point at being at a higher resolution than you can see). Multiple displays would need multiple WIND files but the processing module (eg v.to.rast) wouldn't know which was the appropriate one to use. Thus computational region refers to what is stored in the mapset's WIND file. (sorry gui programmers if I have made small mistakes in that) In gis.m, from the Map Display window if you click on the magnifying glass icon (with the map) you get a pulldown menu. the last item is "set computational region extents to match display". And, the reverse: Zoom display to computational region. Maybe your v.to.rast worked as it should, but you just don't see the correct region in the Map Display ? BTW: I set the Location in GRASS's opening welcome screen, using the Contour.shp file to provide the boundaries. and g.region -p projection: 0 (x,y) You are aware that this means that the Contour.shp did not contain any projection info and that GRASS thus consideres it as unprojected ? Moritz ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from inter secting shapefiles
georgew wrote: > Hamish, please forgive my ignorance but I am not sure what you mean. > As far as I know g.region is the means to change the resolution, correct, > whether through the command line or the GUI. Can you please explain how > to access and change the computational region settings. I was talking about when using it from the GUI. from the command line the display (d.mon) resolution is the computation region. in the GUI there are two: the active display region(+resolution) and the computational one. What you see in the display window(s) is independent of what's in the $MAPSET/WIND file. That WIND(ow) file holds the region which will be used by any modules running. The reason for all this confusion is that you can have multiple GUI display windows open at the same time, each independently zoomed to a different area and set at a different resolution (for a visual display there's no point at being at a higher resolution than you can see). Multiple displays would need multiple WIND files but the processing module (eg v.to.rast) wouldn't know which was the appropriate one to use. Thus computational region refers to what is stored in the mapset's WIND file. (sorry gui programmers if I have made small mistakes in that) In gis.m, from the Map Display window if you click on the magnifying glass icon (with the map) you get a pulldown menu. the last item is "set computational region extents to match display". also in the GIS manager menu there is Config->Region. In that is "Display region settings"* and "Change region settings" which brings up the g.region GUI as if you were running it at the command prompt. [*] perhaps badly worded, nothing to do with the Map Display- it prints the current computational region to the output window. A symptom of the computational region being way out of whack compared to your display region is that you see the result as a series of huge blocks (computational resolution is too coarse), or in the other direction the resolution is so fine that takes ages to load. YMMV hope it helps, Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from inter secting shapefiles
[please don't post in html, it makes it hard to read] georgew wrote: > PS I tried to play around with resolution in the Region dialog but the > display never changed from the same chunky coloured squares. you need to change the computational region, not just the display region settings. the computational region is what the processing modules will use. Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from i ntersecting shapefiles
georgew wrote: Micha, I followed your instructions with the following results: Import data from 2 shapefiles, Contour (layer 1)and Track (layer 2): (no problems reported) v.in.ogr dsn=/home/george/GRASSDATA/ output=testvectmap min_area=0.0001 snap=-1 How many shapefiles did you have in the GRASSDATA directory? What does testvectmap contain? You should make two separate GRASS vectors, one for the tracks and one for the contours. You can use the 'layer' parameter each time to specify each shapefile, like: v.in.ogr dsn=/home/george/GRASSDATA/ layer=contours.shp output=contours Display vector map: (no problems, map displayes correctly) d.vect [EMAIL PROTECTED] color=0:0:0 lcolor=0:0:0 fcolor=170:170:170 display=shape type=point,line,boundary,area icon=basic/circle size=5 layer=-1 lsize=8 xref=left yref=center llayer=1 At this point I'd check to make sure the elevation data came thru: v.info -c # shows the column headers, to make sure you have ELEVATION v.db.select # shows the actual ELEVATION values Convert vector to raster and thin: (no problems reported) v.to.rast [EMAIL PROTECTED] output=testrastmap use=attr type=point,line,area layer=1 column=ELEVATION value=1 rows=10 r.thin [EMAIL PROTECTED] output=thinrastmap iterations=200 Display raster map (PROBLEM:only displays large, brightly coloured rectangles, no contours d.rast [EMAIL PROTECTED] -o How did you set up the Location/Mapset ? You might have a problem with the resolution. What do you get from: r.info testrastmap g.region -p At this point I stopped as I got the impression something was already wrong but could not work out what. Also, should I not rasterize the track data as well? Any help would be appreciated. If you require the source shapefiles I have no problems sending them to you. Let's keep stabbing at it from your end. That's always the best way to understand the methodology. Micha Silver wrote: Yes. I think you should be looking at converting the shapefile contour lines to a raster v.in.org to import the contour lines and tracks into grass v.to.rast and r.thin to convert the contour lines to a raster map. r.surf.contour to create an elevation surface from the rasterized contours r.profile to get the elevations along the tracks. Post back with specific questions if you hit a snag. Regards, Micha ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Elevation profile from i ntersecting shapefiles
georgew wrote: What, no takers? Just a Yes or No will be already of great help to me, Yes. I think you should be looking at converting the shapefile contour lines to a raster, then make an elevation surface from that, and use the r.profile module to calculate elevation profiles for the track lines. So: v.in.org to import the contour lines and tracks into grass v.to.rast and r.thin to convert the contour lines to a raster map. r.surf.contour to create an elevation surface from the rasterized contours r.profile to get the elevations along the tracks. Post back with specific questions if you hit a snag. Regards, Micha anything more, like some hint on a possible solution would be cream on top. Thanks all the same for reading this. George georgew wrote: Hi, I am totally new to GRASS and overwhelmed by its scope and not sure whether the following is possible: I have a set of topographic maps in shapefile format containing contours and walking tracks. The track data does not include elevation (only lat. and long.), but the contour lines do include elevation. Is it possible (and how?) to create an elevation profile for the length of the walking track using the intersecting points between the track and the contour lines to obtain the elevation values? Many thanks for any assistance you can give me. George ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
