[gdal-dev] Re: gdal to define projection, EPSG:3412

John Callahan john.callahan at UDel.Edu
Wed Nov 11 09:06:43 EST 2009


Thanks for responding Hermann. Yes, I tried using -s_srs EPSG:3412 as 
well as the full PROJ4 string with the same result. When specifying 
-s_srs, does gdalwarp ignore (override) the project information in the 
hdr file?

If I include the Hughes 1980 ellipsoid in the hdr file (projection info 
line as described below), then gdalwarp returns that this ellipsoid is 
unrecognized and defaults to WGS84.  It then uses the WGS84 a,b 
parameters and not what I have set through -s_srs.   Well, at least 
that's what returned to the screen with --debug on.  Can I edit some 
config files to allow Hughes 1980 to be there?   (I'm using FWTools 
2.4.6, GDAL 1.7.0dev) 


So, I definitely would like to get my data projected correctly, but I 
also want to make sure my hdr file is correct.  I emailed NSIDC (where I 
obtained the data) to ask for a n appropriate header file.  If I know 
that is correct, then I can focus on the software doing the projecting. 

- John

**************************************************
John Callahan, Geospatial Application Developer
Delaware Geological Survey, University of Delaware
URL: http://www.dgs.udel.edu
**************************************************



Hermann Peifer wrote:
> Did you try: gdalwarp -s_srs EPSG:3412 ?
>
> Hermann
>
> John Callahan wrote:
>> Is it possible to define/set a coordinate system using GDAL?   I 
>> downloaded some data (from NSIDC) I know is in EPSG:3412.  
>> (http://spatialreference.org/ref/epsg/3412/, 
>> http://nsidc.org/data/atlas/epsg_3412.html)
>>
>> The data is in bsq format, therefore requires a hdr file that I've 
>> been piecing together manually.  The hdr file has a "projection info" 
>> line that defines the projection parameters.   Here is the hdr file 
>> I've been using, unsuccessfully:
>>
>> ENVI
>> description = {
>>  File Imported into ENVI.}
>> samples = 316
>> lines   = 332
>> bands   = 1
>> header offset = 300
>> file type = ENVI Standard
>> data type = 1
>> interleave = bsq
>> sensor type = Unknown
>> byte order = 0
>> map info = {NSIDC Sea Ice Polar Stereographic South, 1.0000, 1.0000, 
>> -3950000.0000, 4350000.0000, 2.5000000000e+004, 2.5000000000e+004, 
>> Hughes 1980, units=Meters}
>> projection info = {31, 6378273.0, 6356889.4, -70.000000, 0.000000, 
>> 0.0, 0.0, Hughes1980, NSIDC Sea Ice Polar Stereographic South, 
>> units=Meters}
>> wavelength units = Unknown
>> pixel size = {2.50000000e+004, 2.50000000e+004, units=Meters}
>>
>>
>>
>> The data displays properly with other vector data in EPSG:3412, say 
>> in QGIS.  To me, this simply means the "map info" line of correct.  
>> However, when I try to project the data (gdal_translate, gdalwarp), 
>> the output is nonsense (nearly all null or nodata).  I believe my 
>> "projection info" line is wrong, and may have something to do with 
>> denoting the lat of origin as -90 and standard parallel as -70.   
>> Thanks for any help.
>>
>> - John
>>
>> **************************************************
>> John Callahan, Geospatial Application Developer
>> Delaware Geological Survey, University of Delaware
>> URL: http://www.dgs.udel.edu
>> **************************************************
>
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev


More information about the gdal-dev mailing list