[GRASS-user] what is requried when import data to grass?

When import some raster data to grass,I use the following cmd:

  1. g.region -d
  2. r.in.gdal input=D:/test.TIF output=testTT --overwrite
    I got the error message as following:

Projection of dataset does not appear to match current location.

Location PROJ_INFO is:

name: UTM

datum: nad27

nadgrids: conus

proj: utm

ellps: clark66

a: 6378206.4000000004

es: 0.0067686580

f: 294.9786982000

zone: 13

Import dataset PROJ_INFO is:

cellhd.proj = 0 (unreferenced/unknown)

  1. change the cmd to " r.in.gdal -o input=D:/test.TIF output=testTT --overwrite " I know this override the projection,however I am not sure the potential effect to the data.

Now the tiff is imported,and I got testTT.bule testTT.red testTT.green, then I open them can only see blank. So I wonder it is caused by the projection or location,I use the spanish data as the default location , and I tried to create new location and mapset but never success…

maven apache:

When import some raster data to grass,I use the following cmd:
1) g.region -d

This sets to the "default" region which one previously needs to define,
i.e. needs to create with "g.region -s". Please check the manual of
g.region.

2) r.in.gdal input=D:/test.TIF output=testTT --overwrite

I got the error message as following:
---------------
Projection of dataset does not appear to match current location.
Location PROJ_INFO is:
name: UTM
datum: nad27

[...]

Import dataset PROJ_INFO is:
cellhd.proj = 0 (unreferenced/unknown)
------------------------

That's seems normal. Your test.TIF either isn't georeferenced or it lost
its geotags at some point (after editing with non-GIS tool perhaps?).

3) change the cmd to " r.in.gdal -o input=D:/test.TIF output=testTT
--overwrite " I know this override the projection,however I am not
sure the potential effect to the data.

You should use that only if you _know_ that your test.TIF data are
projected in the system that your grass-location is defined (e.g.
"datum: nad27" as one can see in the error message above). If your data
(test.TIF) are projected in another proj. system expect errors.

Now the tiff is imported,and I got testTT.bule testTT.red
testTT.green, then I open them can only see blank. So I wonder it is
caused by the projection or location,I use the spanish data as the
default location , and I tried to create new location and mapset but
never success...

So, given that test.TIF is in "UTM, datum: nad27" then you need to:
1. import with r.in.gdal -o (as you already did).
2. g.region rast=testTT.blue
3. view your data

If the test.TIF data are projected in another system you need to:
1. find out which is that projection system
2. use some gdal tools to assign the proj. information in the test.TIF
file
3. define a "new" grass-location based on the projection information of
the test.TIF data
4. import with r.in.gdal (without the use of "-o" flag)
5. g.region rast=OneOfTheImportedBands
6. view, process, etc.

Greetings, Nikos

Hi Nikos:
Thanks for your detailed reply!
And I know the reason of the problem now.
However as what you said,I should create a new location and mapset if necessary,it does not work.
I create two folders nLocation/nmapset and create a wind file in the nmapset dir,however when I run the grass gui,I also can not seclect the new location to enter. It throw the error message:the selection location [nlocation ]is not valide. Is the procedure I create the new location is wrong? I followed the maual

maven

2009/6/11 Nikos Alexandris <nikos.alexandris@felis.uni-freiburg.de>

maven apache:

When import some raster data to grass,I use the following cmd:

  1. g.region -d

This sets to the “default” region which one previously needs to define,
i.e. needs to create with “g.region -s”. Please check the manual of
g.region.

  1. r.in.gdal input=D:/test.TIF output=testTT --overwrite

I got the error message as following:

Projection of dataset does not appear to match current location.
Location PROJ_INFO is:
name: UTM
datum: nad27

[…]

Import dataset PROJ_INFO is:
cellhd.proj = 0 (unreferenced/unknown)

That’s seems normal. Your test.TIF either isn’t georeferenced or it lost
its geotags at some point (after editing with non-GIS tool perhaps?).

  1. change the cmd to " r.in.gdal -o input=D:/test.TIF output=testTT
    –overwrite " I know this override the projection,however I am not
    sure the potential effect to the data.

You should use that only if you know that your test.TIF data are
projected in the system that your grass-location is defined (e.g.
“datum: nad27” as one can see in the error message above). If your data
(test.TIF) are projected in another proj. system expect errors.

Now the tiff is imported,and I got testTT.bule testTT.red
testTT.green, then I open them can only see blank. So I wonder it is
caused by the projection or location,I use the spanish data as the
default location , and I tried to create new location and mapset but
never success…

So, given that test.TIF is in “UTM, datum: nad27” then you need to:

  1. import with r.in.gdal -o (as you already did).
  2. g.region rast=testTT.blue
  3. view your data

If the test.TIF data are projected in another system you need to:

  1. find out which is that projection system
  2. use some gdal tools to assign the proj. information in the test.TIF
    file
  3. define a “new” grass-location based on the projection information of
    the test.TIF data
  4. import with r.in.gdal (without the use of “-o” flag)
  5. g.region rast=OneOfTheImportedBands
  6. view, process, etc.

Greetings, Nikos

maven:

And I know the reason of the problem now.

Which is?

However as what you said,I should create a new location and mapset if
necessary,it does not work.
I create two folders nLocation/nmapset and create a wind file in the
nmapset dir, however when I run the grass gui,I also can not seclect
the new location to enter.

Sorry, I never create manually those directories and the WIND file. I
always use the "recommended" way, that is through GRASS.

Read _A)_ or/ and _B)_ in:
http://grass.osgeo.org/grass64/manuals/html64_user/helptext.html

It throw the error message:the selection location [nlocation ]is not
valide. Is the procedure I create the new location is wrong? I
followed the maual

Which manual?

Nikos

Hi:
I followed the doc from the web resrouce:
https://svn.wald.intevation.org/svn/pywps/tags/pywps-3.1.0/pywps/Grass.py

And I think what we are doing is similar to theirs,so I thought the confignation of grass maybe correct for us.

2009/6/11 Nikos Alexandris <nikos.alexandris@felis.uni-freiburg.de>

maven:

And I know the reason of the problem now.

Which is?

However as what you said,I should create a new location and mapset if
necessary,it does not work.
I create two folders nLocation/nmapset and create a wind file in the
nmapset dir, however when I run the grass gui,I also can not seclect
the new location to enter.

Sorry, I never create manually those directories and the WIND file. I
always use the “recommended” way, that is through GRASS.

Read A) or/ and B) in:
http://grass.osgeo.org/grass64/manuals/html64_user/helptext.html

It throw the error message:the selection location [nlocation ]is not
valide. Is the procedure I create the new location is wrong? I
followed the maual

Which manual?

Nikos

maven apache wrote:

Hi:
I followed the doc from the web resrouce:
https://svn.wald.intevation.org/svn/pywps/tags/pywps-3.1.0/pywps/Grass.py

And I think what we are doing is similar to theirs,so I thought the confignation of grass maybe correct for us.

Is there some special reason you need to create the GRASS directory structure programatically each time? Can’t you just start GRASS once the straight-forward way to make the required location, mapset and files, then use the instructions here:
http://grass.osgeo.org/wiki/GRASS_and_Shell
to run your grass commands from a shell/python script?

Is there some special reason you need to create the GRASS directory structure programatically each time? Can’t you just start

As the pywps I have refered,each client(user) may send their own process request with different map,obvirous they do not have the save location or mapset even the projection,so if config the grass only once ,do the location and mapset accord with the user requested map?

GRASS once the straight-forward way to make the required location, mapset and files, then use the instructions here:
http://grass.osgeo.org/wiki/GRASS_and_Shell
to run your grass commands from a shell/python script?

I read the shell,actually I am using another script in the annex,and I think they have the same thought.

2009/6/11 Micha Silver <micha@arava.co.il>

maven apache wrote:

Hi:
I followed the doc from the web resrouce:
https://svn.wald.intevation.org/svn/pywps/tags/pywps-3.1.0/pywps/Grass.py

And I think what we are doing is similar to theirs,so I thought the confignation of grass maybe correct for us.

Is there some special reason you need to create the GRASS directory structure programatically each time? Can’t you just start GRASS once the straight-forward way to make the required location, mapset and files, then use the instructions here:
http://grass.osgeo.org/wiki/GRASS_and_Shell
to run your grass commands from a shell/python script?


Micha

2009/6/11 Nikos Alexandris <nikos.alexandris@felis.uni-freiburg.de>

maven:

And I know the reason of the problem now.

Which is?

However as what you said,I should create a new location and mapset if
necessary,it does not work.
I create two folders nLocation/nmapset and create a wind file in the
nmapset dir, however when I run the grass gui,I also can not seclect
the new location to enter.

Sorry, I never create manually those directories and the WIND file. I
always use the “recommended” way, that is through GRASS.

Read A) or/ and B) in:
http://grass.osgeo.org/grass64/manuals/html64_user/helptext.html

It throw the error message:the selection location [nlocation ]is not
valide. Is the procedure I create the new location is wrong? I
followed the maual

Which manual?

Nikos

This mail was received via Mail-SeCure System.


---

_______________________________________________
grass-user mailing list
[grass-user@lists.osgeo.org](mailto:grass-user@lists.osgeo.org)
[http://lists.osgeo.org/mailman/listinfo/grass-user](http://lists.osgeo.org/mailman/listinfo/grass-user)

This mail was received via Mail-SeCure System.

  
(attachments)

Grass.java (12.2 KB)

maven apache wrote:

    Is there some special reason you need to create the GRASS
    directory structure programatically each time? Can't you just start

As the pywps I have refered,each client(user) may send their own process request with different map,obvirous they do not have the save location or mapset even the projection,so if config the grass only once ,do the location and mapset accord with the user requested map?

Well, no. Each location has a specific coordinate reference system, set up when you create the location. Any data pulled into that location should match that CRS.
I don't recall the start of your thread: Are you saying that you need to be able to handle any data that clients throw at you in *any* CRS? How will you know the CRS of the client's maps? (Many data formats do not require full metadata with CRS information)

And one other point came to my mind: be aware that the user calling the GRASS script must be the owner of the $GISDBASE directory structure. So if these are web based apps, then the user running your http server must own the GRASS database directory.

Good luck, and keep us posted,
--
Micha

    GRASS once the straight-forward way to make the required location,
    mapset and files, then use the instructions here:
    http://grass.osgeo.org/wiki/GRASS_and_Shell
    to run your grass commands from a shell/python script?

I read the shell,actually I am using another script in the annex,and I think they have the same thought.
2009/6/11 Micha Silver <micha@arava.co.il <mailto:micha@arava.co.il>>

    maven apache wrote:

    Hi:
    I followed the doc from the web resrouce:
    https://svn.wald.intevation.org/svn/pywps/tags/pywps-3.1.0/pywps/Grass.py
    And I think what we are doing is similar to theirs,so I thought
    the confignation of grass maybe correct for us.

    Is there some special reason you need to create the GRASS
    directory structure programatically each time? Can't you just
    start GRASS once the straight-forward way to make the required
    location, mapset and files, then use the instructions here:
    http://grass.osgeo.org/wiki/GRASS_and_Shell
    to run your grass commands from a shell/python script?

    -- Micha
     

    2009/6/11 Nikos Alexandris
    <nikos.alexandris@felis.uni-freiburg.de
    <mailto:nikos.alexandris@felis.uni-freiburg.de>>

        maven:
        > And I know the reason of the problem now.

        Which is?

        > However as what you said,I should create a new location and
        mapset if
        > necessary,it does not work.
        > I create two folders nLocation/nmapset and create a wind
        file in the
        > nmapset dir, however when I run the grass gui,I also can
        not seclect
        > the new location to enter.

        Sorry, I never create manually those directories and the WIND
        file. I
        always use the "recommended" way, that is through GRASS.

        Read _A)_ or/ and _B)_ in:
        http://grass.osgeo.org/grass64/manuals/html64_user/helptext.html

        > It throw the error message:the selection location
        [nlocation ]is not
        > valide. Is the procedure I create the new location is wrong? I
        > followed the maual

        Which manual?

        Nikos

    This mail was received via Mail-SeCure System.
    ------------------------------------------------------------------------

    _______________________________________________
    grass-user mailing list
    grass-user@lists.osgeo.org <mailto:grass-user@lists.osgeo.org>
    http://lists.osgeo.org/mailman/listinfo/grass-user

    This mail was received via Mail-SeCure System.

This mail was received via Mail-SeCure System.

I don’t recall the start of your thread:

Are you saying that you need to be able to handle any data that clients throw at you in any CRS?

How will you know the CRS of the client’s maps? (Many data formats do not require full metadata with CRS information)

In fact I perfer to create location and mapset for each requested map according the crs or other parameter, and how to get the parameter is another topic( for some ogc web servcie wcs wms we can get most paramter from the requsted url for example:
http://localhost:8080/wcs?Request=wcs&identifier=imageID&crs=EPSG:4326&bbox=[][]$format=tiff…)
So what block me is the location and mapset, in a word I am so unfamiliar with the grass although I have read some documents that I can not confignation as the pywps did.

And one other point came to my mind: be aware that the user calling the GRASS script must be the owner of the $GISDBASE directory structure. So if these are web based apps, then the user running your http server must own the GRASS database directory.

What we do is to provide a interface to the user, the process name and some required parameter for the grass, the real requestor of the grass is ourselvers , in other words it is the server (we own the GISDBASE)to call the grass,and the parameter passed to grass come from the user.
May be I can describe it as follows:
the user(web browse) provide the paramter and the original map ------->our server(get the para and the map) -----> call the grass (pass the praa and map) ----->grass process----->grass send the result,(may be exported) ----->server get the requested data —> send the request data to user client

Good luck, and keep us posted,

Micha

GRASS once the straight-forward way to make the required location,
mapset and files, then use the instructions here:
http://grass.osgeo.org/wiki/GRASS_and_Shell
to run your grass commands from a shell/python script?

I read the shell,actually I am using another script in the annex,and I think they have the same thought.

2009/6/11 Micha Silver <micha@arava.co.il mailto:[micha@arava.co.il](mailto:micha@arava.co.il)>

maven apache wrote:

Hi:
I followed the doc from the web resrouce:
https://svn.wald.intevation.org/svn/pywps/tags/pywps-3.1.0/pywps/Grass.py
And I think what we are doing is similar to theirs,so I thought
the confignation of grass maybe correct for us.

Is there some special reason you need to create the GRASS
directory structure programatically each time? Can’t you just
start GRASS once the straight-forward way to make the required
location, mapset and files, then use the instructions here:
http://grass.osgeo.org/wiki/GRASS_and_Shell
to run your grass commands from a shell/python script?

– Micha

2009/6/11 Nikos Alexandris
<nikos.alexandris@felis.uni-freiburg.de

mailto:[nikos.alexandris@felis.uni-freiburg.de](mailto:nikos.alexandris@felis.uni-freiburg.de)>

maven:

And I know the reason of the problem now.

Which is?

However as what you said,I should create a new location and
mapset if
necessary,it does not work.
I create two folders nLocation/nmapset and create a wind
file in the
nmapset dir, however when I run the grass gui,I also can
not seclect
the new location to enter.

Sorry, I never create manually those directories and the WIND
file. I
always use the “recommended” way, that is through GRASS.

Read A) or/ and B) in:
http://grass.osgeo.org/grass64/manuals/html64_user/helptext.html

It throw the error message:the selection location
[nlocation ]is not
valide. Is the procedure I create the new location is wrong? I
followed the maual

Which manual?

Nikos

This mail was received via Mail-SeCure System.


grass-user mailing list

grass-user@lists.osgeo.org mailto:[grass-user@lists.osgeo.org](mailto:grass-user@lists.osgeo.org)

http://lists.osgeo.org/mailman/listinfo/grass-user

This mail was received via Mail-SeCure System.

This mail was received via Mail-SeCure System.

maven apache wrote:

    I don't recall the start of your thread:

    Are you saying that you need to be able to handle any data that
    clients throw at you in *any* CRS?

    How will you know the CRS of the client's maps? (Many data formats
    do not require full metadata with CRS information)

In fact I perfer to create location and mapset for each requested map according the crs or other parameter, and how to get the parameter is another topic( for some ogc web servcie wcs wms we can get most paramter from the requsted url for example:
http://localhost:8080/wcs?Request=wcs&identifier=imageID&crs=EPSG:4326&bbox=[][]$format=tiff <http://localhost:8080/wcs?Request=wcs&identifier=imageID&crs=EPSG:4326&bbox=[][]$format=tiff&gt;\.\.\.\.\.\)
So what block me is the location and mapset, in a word I am so unfamiliar with the grass although I have read some documents that I can not confignation as the pywps did.

Here's a test script that I can run (pretty much copied from the wiki http://grass.osgeo.org/wiki/GRASS_and_Shell ) It requires GRASS 6.4.

#!/bin/bash
# Example in bash shell syntax:
# path to GRASS binaries and libraries:
export GISBASE=/usr/local/grass64/grass-6.4.0RC4/
export PATH=$PATH:$GISBASE/bin:$GISBASE/scripts
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$GISBASE/lib
# use process ID (PID) as lock file number:
export GIS_LOCK=$$
# path to GRASS settings file
export GISRC=$HOME/.grassrc6
export GISDBASE=$HOME/geodata/grass
export LOCATION_NAME=testloc
export MAPSET=testmap

####################
# Here is my stuff
# Create new location and mapset in EPSG:2039 CRS
g.proj -c loc=$LOCATION_NAME epsg=2039
g.mapset -c --quiet map=$MAPSET loc=$LOCATION_NAME gisdbase=$GISDBASE
echo "200000,500000" | v.in.ascii out=testpoint fs=, --quiet
v.info testpoint
#######################

#run GRASS' cleanup routine
$GISBASE/etc/clean_temp
# remove session tmp directory:
rm -rf /tmp/grass6-$USER-$GIS_LOCK

HTH,

Hi:

I'm cc'ing the GRASS list, since someone else may have additional ideas.

maven apache wrote:

Hi:
I have set all the environment variables needed by Grass,so I start run the grass commond.
********Here I have set the env and the g in the "g,run" respent a java class to config the grass**********
  g.run("r.info <http://r.info> map=roads"); // here I intent to test whether I have enter the default grass location(spanish60) or not
  g.run("g.proj -c georef=d:/kk.tif datumtrans=0 location=mainLocation"); // create new location and mapset accord a tiff
  g.setLocationAndMapSet("mainLocation", "PERMANENT"); //reset the locaton and mapset
  g.run("r.in.gdal input=d:/kk.tif out=kkTest");

Most likely you will want to add here: g.region rast=kkTest
This insures that the computational region is the same as the imported tiff.
If you don't set the region to the raster's extent then you'll get the "black blank" image you mentioned below.

  g.run("r.info <http://r.info> map=kkTest");
  g.run("r.buffer input=kkTest output=kkout distances=100 units=meters");
  g.run("r.out.gdal input=kkout format=GTiff output=d:/kkgeotiffTest.tif"); //here I got a black blank image which I can not open it with common image reader
  g.run("r.out.tiff input=kkout output=d:/kktiffTest.tif compression=none --quiet"); //here I got a buffered image I can see the roads in it with the commo image viewer. I am not sure why? They are all exported to tiff just by two methods..

As far as I know, r.out.tiff is just a simplified version of r.out.gdal. When you use r.out.gdal you have more options available, In particular there's the "createopt" option, So you can do:
r.out.gdal in=kkout out=d:/kktiffTest,tif createopt="INTERLEAVE=PIXEL,TFW=YES"

The setLocationAndMapSet method:
public void setLocationAndMapSet(String nLocation,String nMapset) {
  String gisRc="grassrc";
  FileWriter fstream;
  try {
   fstream = new java.io.FileWriter(gisRc);
   BufferedWriter out = new java.io.BufferedWriter(fstream);
   out.write("LOCATION_NAME: " + nLocation + " \n");
   out.write("GISDBASE: " + grassDataBase + "\n");
   out.write("MAPSET: " + nMapset + " \n");
   out.write("DIGITIZER: none\n");
   out.write("OVERWRITE: 1\n");
   out.write("GRASS_GUI: text\n");
   out.close();
  } catch (IOException e) {
   // TODO Auto-generated catch block
   e.printStackTrace();
  }
  this.setEnv(grassPath, null,gisRc,nMapset, nLocation, grassDataBase,grassPath+File.separator+"proj");
   }
In any case ,I have make a progrss, :slight_smile:
I have some extra questions:
1) take the using grass gui for instance,I want to know if user can change its location dynamically or he can only change it by restart the grass ? I just want to know the above method is improveable?

g.mapset is the command to change the location or mapset within a GRASS session

2) the differedt resut of two differente export way( r.out.gdal r.out.tiff)

See above.

3) Can I delete the location using some grass commond? I mean I want to delete the created location and mapset after a user quested is completed.

Just delete the whole location directory.

Thanks!

Regards,
Micha

2009/6/14 Micha Silver <micha@arava.co.il <mailto:micha@arava.co.il>>

    maven apache wrote:

         I think you should start this way, and

        work out the string of GRASS commands you will need *before*
        you start scripting.

    Sorry but I am not exactly know your meaning.

    I just suggest you start by working out the GRASS commands you
    will need from *within GRASS*. For example, for the script I sent
    you earlier, I first started GRASS, then I ran those commands, one
    by one. This way I can check that the commands are working as I
    want. Then, I copied them into the script and, after exiting
    GRASS, I ran the script.

        Then, once you have the workflow running correctly, you can,

        from outside of GRASS,

    Also,what?

    It's very possible to run GRASS commands without the GRASS
    interface, but you first have to set up the environment: some
    environ variables and path. The basic example script is a demo how
    it could be done. Once you have setup those path and environment
    variables, you can run a script of GRASS commands without starting
    the interface.
    Hope that's clearer now...
    -- Micha

        setup the environment and run that same workflow.

    This mail was received via Mail-SeCure System.

This mail was received via Mail-SeCure System.

Micha Silver wrote:

As far as I know, r.out.tiff is just a simplified version of r.out.gdal.

No.
r.out.tiff exports an image that looks like what you see in the display,
in other words, it translates the color rules of a GRASS raster to a
tiff image. The cell values of the GRASS raster are *not* exported by
r.out.tiff.
r.out.gdal exports raster data (cell values) and, when possible, embeds
color rules in the metadata. The exported data may or may not look like
the GRASS display of the raster, depending on the data type and file
format choosen for export and the capabilities of the spatial data
viewer used to display the exported data.

Markus M

Can I understand the difference as the difference between the geotiff and the tiff?

2009/6/15 Markus GRASS <markus.metz.giswork@googlemail.com>

Micha Silver wrote:

As far as I know, r.out.tiff is just a simplified version of r.out.gdal.

No.
r.out.tiff exports an image that looks like what you see in the display,
in other words, it translates the color rules of a GRASS raster to a
tiff image. The cell values of the GRASS raster are not exported by
r.out.tiff.
r.out.gdal exports raster data (cell values) and, when possible, embeds
color rules in the metadata. The exported data may or may not look like
the GRASS display of the raster, depending on the data type and file
format choosen for export and the capabilities of the spatial data
viewer used to display the exported data.

Markus M

maven apache wrote:

Can I understand the difference as the difference between the geotiff
and the tiff?

No. A geotiff can be a tiff with georeferencing information (projection,
extends, resolution). The difference between r.out.tiff output and
r.out.gdal output is what kind of information is exported: color rules
for r.out.tiff, cell values for r.out.gdal. As an example, use some
elevation data in one of the sample datasets, set different color rules,
export each time with r.out.tiff and with r.out.gdal. r.out.tiff output
(pixel values) is dependent on the current color rules, r.out.gdal
output (pixel values) stays the same (as long as you don't change the
region), but the embedded color table may change.

2009/6/15 Markus GRASS <markus.metz.giswork@googlemail.com
<mailto:markus.metz.giswork@googlemail.com>>

    Micha Silver wrote:
    > As far as I know, r.out.tiff is just a simplified version of
    r.out.gdal.
    No.
    r.out.tiff exports an image that looks like what you see in the
    display,
    in other words, it translates the color rules of a GRASS raster to a
    tiff image. The cell values of the GRASS raster are *not* exported by
    r.out.tiff.
    r.out.gdal exports raster data (cell values) and, when possible,
    embeds
    color rules in the metadata. The exported data may or may not look
    like
    the GRASS display of the raster, depending on the data type and file
    format choosen for export and the capabilities of the spatial data
    viewer used to display the exported data.

    Markus M

------------------------------------------------------------------------

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user
  

Markus GRASS wrote:

Micha Silver wrote:
  

As far as I know, r.out.tiff is just a simplified version of r.out.gdal.
    

No.
r.out.tiff exports an image that looks like what you see in the display,
in other words, it translates the color rules of a GRASS raster to a
tiff image. The cell values of the GRASS raster are *not* exported by
r.out.tiff.
r.out.gdal exports raster data (cell values) and, when possible, embeds
color rules in the metadata. The exported data may or may not look like
the GRASS display of the raster, depending on the data type and file
format choosen for export and the capabilities of the spatial data
viewer used to display the exported data.

Markus M

This mail was received via Mail-SeCure System.

Thanks Markus,
--
Micha