Hi everybody-
I need a little bit assistance in projection stuff, importing
raster files etc.
What I have:
Different kind of datasets:
1) passive microwave, gridded on a regular grid as shown in the figure.
(it is the SSM/I - NSIDC grid)
2) optical and infrared raw data (AVHRR)
3) gtopo30 topography
What I did:
Set up a GRASS5 mapset which fits the SSM/I grid; it is an x-y mapset.
What I want:
Import datatype 2) and 3) into this mapset. 
In case of the AVHRR data I am able to choose from a varity of
projections
in which I want to have the raw data after some basic processing, but
there
are "only" geographic ones (like polar stereographic, which I thought it
would be most suitable, since the SSM/I grid is based on a polar
stereographic grid - unfortunatly, I am not able to get a dataset with
the
same edges as the mapset - it is always of the shape of trapecoid (sp?))
If I know how to import the gtopo30 into my mapset I will be able to
manage
the AVHRR data since gtopo30 is in "Equidistant Cylindrical" - I can get
the
AVHRR data into a similar grid...
lat = -39.23 deg lat = -39.23 deg
lon = 317.76 deg lon = 42.24 deg
+----------- 4350 -----------+
| |
| |
| |
| |
| |
-3950 o 3950
| Southpole |
| |
| |
| |
| |
| |
+---------- -3950 -----------+
lat = -41.45 deg lat = -41.45 deg
lon = 225.00 deg lon = 135.00 deg
dx = 12.5
dy = 12.5
(A (slightly) better image is available at
http://www-nsidc.colorado.edu/NASA/GUIDE/SSMI/gridmap.JPG
on the right side - even though the anotations of the upper right and
the
lower left corner have to be changed...)
What I don't know:
is how to "jump" between / fit / take account to the different types of
projections (how ever you want to call it)...
Any suggestions will be highly appriciated!!!
--
------------------------------------------------------------------------
Stephan Eickschen Tel. : +49 (0)251 83-34704
Westfaelische Wilhelms-Universitaet Fax. : +49 (0)251 83-36100
Institute for Geophysics
Research Unit of Physical Glaciology email: eicksch@uni-muenster.de
Corrensstrasse 24
48149 Muenster
Germany
http://earth.uni-muenster.de/~eicksch/
------------------------------------------------------------------------
Could'nt you use the proj package by Evenden? It used to be part
of the grass garden.
Agus
****NOTE NEW PHONE NUMBER*****
Dr. Agustin Lobo
Instituto de Ciencias de la Tierra (CSIC)
Lluis Sole Sabaris s/n
08028 Barcelona SPAIN
tel 34 93409 5410
fax 34 93411 0012
alobo@ija.csic.es
http://pangea.ija.csic.es/alobo
On Wed, 10 Feb 1999, Stephan Eickschen wrote:
Hi everybody-
I need a little bit assistance in projection stuff, importing
raster files etc.
What I have:
Different kind of datasets:
1) passive microwave, gridded on a regular grid as shown in the figure.
(it is the SSM/I - NSIDC grid)
2) optical and infrared raw data (AVHRR)
3) gtopo30 topography
What I did:
Set up a GRASS5 mapset which fits the SSM/I grid; it is an x-y mapset.
What I want:
Import datatype 2) and 3) into this mapset.
In case of the AVHRR data I am able to choose from a varity of
projections
in which I want to have the raw data after some basic processing, but
there
are "only" geographic ones (like polar stereographic, which I thought it
would be most suitable, since the SSM/I grid is based on a polar
stereographic grid - unfortunatly, I am not able to get a dataset with
the
same edges as the mapset - it is always of the shape of trapecoid (sp?))
If I know how to import the gtopo30 into my mapset I will be able to
manage
the AVHRR data since gtopo30 is in "Equidistant Cylindrical" - I can get
the
AVHRR data into a similar grid...
lat = -39.23 deg lat = -39.23 deg
lon = 317.76 deg lon = 42.24 deg
+----------- 4350 -----------+
| |
| |
| |
| |
| |
-3950 o 3950
| Southpole |
| |
| |
| |
| |
| |
+---------- -3950 -----------+
lat = -41.45 deg lat = -41.45 deg
lon = 225.00 deg lon = 135.00 deg
dx = 12.5
dy = 12.5
(A (slightly) better image is available at
http://www-nsidc.colorado.edu/NASA/GUIDE/SSMI/gridmap.JPG
on the right side - even though the anotations of the upper right and
the
lower left corner have to be changed...)
What I don't know:
is how to "jump" between / fit / take account to the different types of
projections (how ever you want to call it)...
Any suggestions will be highly appriciated!!!
--
------------------------------------------------------------------------
Stephan Eickschen Tel. : +49 (0)251 83-34704
Westfaelische Wilhelms-Universitaet Fax. : +49 (0)251 83-36100
Institute for Geophysics
Research Unit of Physical Glaciology email: eicksch@uni-muenster.de
Corrensstrasse 24
48149 Muenster
Germany
http://earth.uni-muenster.de/~eicksch/
------------------------------------------------------------------------
Agus-
Could'nt you use the proj package by Evenden? It used to be part
of the grass garden.
As far as I understand the manual of proj it transforms the "values" of
given coordinates. What I would like / have to do is a transformation /
reprojection of a whole image - it should something similar to
i.rectify without having to find pixels / point in the "source image"
but knowing its projection parameters - you understand???
I am now trying the following: I have created a (new) x-y location
which fits the specification of my image. I successfully imported the
image into this location. Is it now possible to import and reproject it
now into the desired location?
Regards,
Stephan
--
------------------------------------------------------------------------
Stephan Eickschen Tel. : +49 (0)251 83-34704
Westfaelische Wilhelms-Universitaet Fax. : +49 (0)251 83-36100
Institute for Geophysics
Research Unit of Physical Glaciology email: eicksch@uni-muenster.de
Corrensstrasse 24
48149 Muenster
Germany
http://earth.uni-muenster.de/~eicksch/
------------------------------------------------------------------------
As far as I understand the manual of proj it transforms the "values" of
given coordinates. What I would like / have to do is a transformation /
reprojection of a whole image - it should something similar to
i.rectify without having to find pixels / point in the "source image"
but knowing its projection parameters - you understand???
Yes, I do. There are 2 steps:
1. Finding the new coordinates: proj would do it
2. geting the new values of the new pixels by interpolation: normally
nearest neighbour. Check on a RS book the orthorectification process,
i.e. Richards 1993, S-V.
Regarding the actual process, I do not know if grass5. has the necessary
programs to re-project rasters, I guess it does but only between a limited
number of projections. You must either do the processing in a different
package and then import to grass once you have all your imagery in the
same projection, or do the interpolation by yourself in grass. The second
solution would imply:
1. Defining the corners of your image in the new projection.
2. Defining the corners of the new image in the new projection (i.e.,
define the rectangle including the corners calculated in 1.)
3. Defining the new resolution.
4. Defining the centers of the new pixels in the new coordinate system.
5. Calculating the centers of the new pixels in the old coordinate system.
6. Selecting the nearest old pixel for each new pixel (if you follow a
nearest neighbour method) and assigning this value to the new pixel.
In any case, you must know how to pass from your arbitrary x,y coordinate
system to your "Equidistant Cylindrical" projection back and forth.
Normally you'll have to pass from x,y to lon,lat and then to "Equidistant
Cylindrical" coordinates. If you do not know how to pass from
the x,y coordinates to lon,lat or to any projection, you
must use i.rectify to georectify to a known projection, i.e.
UTM. Then reproject your "Equidistant Cylindrical" DEM to UTM.
Personally, I would look for a package
supporting the "Equidistant Cylindrical" and do that work with it
(I guess that grass5.0 does not support "Equidistant Cylindrical"). Then
export to grass.
Agus
