[GRASS-user] problem with disk space. Raster data

Hi,

I'm performing a couple of treatments over a specific region and my
problems arise when I deal with UTM projection in contrast with
Lat/Lon projection.

1) I import my raster data (Aster Dem) using the properties of the
georreferenced GeoTiff image with wgs84 datum and lat/lon projection
2) I perform, for example, a r.shaded.relief over the data and
everything runs fine. I get my new file in a decent time

The problem arise when after creating a newlocation with the
corresponding lat/lon coordinates transformed to UTM ones I try to
follow the steps above.

1) The import of the new data is OK and I can see the image with no
problem, but...
2) performing the r.shaded.relief process I got a full hard disk
(100%) and an overnight run for the computer

Using r.info for both original data sources I get:

A) Lat/Lon Proj

Type of Map: raster Number of Categories: 5988 |
| Data Type: CELL |
| Rows: 3601 |
| Columns: 3601 |
| Total Cells: 12967201 |
| Projection: Latitud - Longitud. |
| N: 7:59:59.5S S: 9:00:00.5S Res: 0:00:01 |
| E: 75:59:59.5W W: 77:00:00.5W Res: 0:00:01 |
| Range of data: min = -580 max = 5988

B) UTM Proj

Type of Map: raster Number of Categories: 6744 |
| Data Type: CELL |
| Rows: 2965 |
| Columns: 2282 |
| Total Cells: 6766130 |
| Projection: UTM (zone -18) |
| N: 9055864.24489094 S: 8964905.38350331 Res: 30.67752492 |
| E: 244988.11769175 W: 174982.00582445 Res: 30.67752492 |
| Range of data: min = 0 max = 6744

As you can notice, the UTM region is even smaller tha the Lat/Lon one. Any idea?

Thanks!
Antonio

PD: I'm running a Debian Intel machine with 2GB of RAM

The problem arise when after creating a newlocation with the
corresponding lat/lon coordinates transformed to UTM ones I try to
follow the steps above.

how did you reproject the data?:

within grass (r.proj) or outside grass (i.e. gdalwarp,...)?

can you post the command for raster-reprojection and g.region -p of the
Lat/Lon-location and the UTM-location.

-----
best regards
Helmut
--
View this message in context: http://osgeo-org.1560.n6.nabble.com/problem-with-disk-space-Raster-data-tp5002151p5002153.html
Sent from the Grass - Users mailing list archive at Nabble.com.

Hi Helmut,
The process was as follows:

  1. I defined a newlocation with the coordinates in lat/lon of the NE and SW corners of the area of the data file, converted to UTM, previously
    defining the EPSG code (in my case 32718)
  2. Once in Grass I imported my data file (Aster Dem, geotiff format) with the r.in.gdal tool and forced the data to use the projection defined in my newlocation. Everything went fine and I am able to visualize my dem with its new UTM coordinates.

Any idea?
Cheers!
Antonio

El domingo, 16 de septiembre de 2012, Helmut Kudrnovsky escribió:

The problem arise when after creating a newlocation with the
corresponding lat/lon coordinates transformed to UTM ones I try to
follow the steps above.

how did you reproject the data?:

within grass (r.proj) or outside grass (i.e. gdalwarp,…)?

can you post the command for raster-reprojection and g.region -p of the
Lat/Lon-location and the UTM-location.


best regards
Helmut

View this message in context: http://osgeo-org.1560.n6.nabble.com/problem-with-disk-space-Raster-data-tp5002151p5002153.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

Antonio wrote:

I'm performing a couple of treatments over a specific region
and my problems arise when I deal with UTM projection in
contrast with Lat/Lon projection.

1) I import my raster data (Aster Dem) using the properties
of the georreferenced GeoTiff image with wgs84 datum and lat/lon
projection
2) I perform, for example, a r.shaded.relief over the data
and everything runs fine. I get my new file in a decent time

The problem arise when after creating a newlocation with
the corresponding lat/lon coordinates transformed to UTM ones I
try to follow the steps above.

1) The import of the new data is OK and I can see the image
with no problem, but...
2) performing the r.shaded.relief process I got a full hard
disk (100%) and an overnight run for the computer

what does 'g.region -p' say about the computational region?

in your utm location try 'g.region rast=YOUR_MAP_NAME_HERE'
before running raster modules to set the computational region
to match that map's bounds and resolution.

Hamish