[gdal-dev] importing HDF sea ice grid with unusual projection

Seb spluque at gmail.com
Thu Nov 5 13:27:10 EST 2009


Hi,

I originally posted to the GRASS list, but it was suggested that it
would be more appropriate here, given that GRASS uses GDAL for this
particular problem.

I'm in the process of importing sea ice concentration data from:

ftp://ftp-projects.zmaw.de/seaice/AMSR-E_ASI_IceConc

I have chosen to use their HDF product because their Geotiff uses a
WGS84 datum that results in up to 500 m error at the borders, according
to their info files at the root of each subdirectory.

The grids are not georeferenced, but they use the same grid that NSIDC
does with the specifications given here:

http://nsidc.org/data/docs/daac/nsidc0002_ssmi_seaice.gd.html

Fortunately, there's an EPSG code (3411, for the northern hemisphere)
for that projection.

gdalinfo on the HDF file:

---<--------------------cut here---------------start------------------->---
$ gdalinfo asi-n6250-20020619-v5i.hdf 
Driver: HDF4Image/HDF4 Dataset
Files: asi-n6250-20020619-v5i.hdf
Size is 1216, 1792
Coordinate System is `'
Metadata:
  valid_range=0, 1
  long_name=ASI Ice Concentration, 20020619, res: 6.25000, AMSR-E, ASI Version: 5.5i, missing data interpolated.
Corner Coordinates:
Upper Left  (    0.0,    0.0)
Lower Left  (    0.0, 1792.0)
Upper Right ( 1216.0,    0.0)
Lower Right ( 1216.0, 1792.0)
Center      (  608.0,  896.0)
Band 1 Block=1216x82 Type=Float32, ColorInterp=Gray
---<--------------------cut here---------------end--------------------->---

Converting to GTiff for importing into GRASS, using the EPSG code and
bounds in the NSIDC's info:

---<--------------------cut here---------------start------------------->---
$ gdal_translate -of GTiff -a_srs EPSG:3411 -a_ullr -3800000 5600000 3800000 -5600000 asi-n6250-20020619-v5i.hdf asi-n6250-20020619-v5i.tif 
Warning 6: A dataset opened by GDALOpenShared should have the same filename (asi-n6250-20020619-v5i.hdf) and description (HDF4_SDS:UNKNOWN:"asi-n6250-20020619-v5i.hdf":0)
Input file size is 1216, 1792
0...10...20...30...40...50...60...70...80...90...100 - done.

$ gdalinfo asi-n6250-20020619-v5i.tif 
Driver: GTiff/GeoTIFF
Files: asi-n6250-20020619-v5i.tif
Size is 1216, 1792
Coordinate System is:
PROJCS["NSIDC Sea Ice Polar Stereographic North",
    GEOGCS["Unspecified datum based upon the Hughes 1980 ellipsoid",
        DATUM["Not_specified_based_on_Hughes_1980_ellipsoid",
            SPHEROID["Hughes 1980",6378273,298.279411123061,
                AUTHORITY["EPSG","7058"]],
            AUTHORITY["EPSG","6054"]],
        PRIMEM["Greenwich",0],
        UNIT["degree",0.0174532925199433],
        AUTHORITY["EPSG","4054"]],
    UNIT["metre",1,
        AUTHORITY["EPSG","9001"]],
    AUTHORITY["EPSG","3411"]]
Origin = (-3800000.000000000000000,5600000.000000000000000)
Pixel Size = (6250.000000000000000,-6250.000000000000000)
Metadata:
  AREA_OR_POINT=Area
  valid_range=0, 1
  long_name=ASI Ice Concentration, 20020619, res: 6.25000, AMSR-E, ASI Version: 5.5i, missing data interpolated.
Image Structure Metadata:
  INTERLEAVE=BAND
Corner Coordinates:
Upper Left  (-3800000.000, 5600000.000) 
Lower Left  (-3800000.000,-5600000.000) 
Upper Right ( 3800000.000, 5600000.000) 
Lower Right ( 3800000.000,-5600000.000) 
Center      (   0.0000000,   0.0000000) 
Band 1 Block=1216x1 Type=Float32, ColorInterp=Gray
---<--------------------cut here---------------end--------------------->---

So far so good, but when importing into a GRASS location:

---<--------------------cut here---------------start------------------->---
r.in.gdal in=/home/sluque/tmp/asi-n6250-20020619-v5i.tif out=test location=Test
WARNING: No projection name! Projection parameters likely to be
         meaningless.
WARNING: Datum <Not_specified_based_on_Hughes_1980_ellipsoid> not
         recognised by GRASS and no parameters found
Location <Test> created
 100%
r.in.gdal complete. Raster map <test> created.
---<--------------------cut here---------------end--------------------->---

What is missing here?  I've also tried using the proj4 strings offered
for that EPSG, but the results are the same.  Any tips would be appreciated.


Cheers,

-- 
Seb



More information about the gdal-dev mailing list