Simon, I skimmed the code and there is no actual datum support in the grib reader. It does attempt to differentiate between ellipsoids and spheres so I assume the file actually did record things as a sphere. Generally speaking it is not clear to me that the GRIB driver can actually do better based on the information in the files but I'd be happy to be proven wrong. In the meantime, it appears you will need to overwrite the geographic coordinate system in your ingestor.
Best regards, Frank On Wed, Mar 11, 2015 at 1:25 PM, Simon (Vsevolod) Ilyushchenko < sim...@google.com> wrote: > Hi, > > If I use GDAL to read GRIB files like these: > > ftp://hydro1.sci.gsfc.nasa.gov/data/s4pa/NLDAS/NLDAS_NOAH0125_H.002/2015/065/NLDAS_NOAH0125_H.A20150306.1300.002.grb > > the CRS I get back is: > GEOGCS["Coordinate System imported from GRIB file", > DATUM["unknown", > SPHEROID["Sphere",6371200,0]], > PRIMEM["Greenwich",0], > UNIT["degree",0.0174532925199433]] > > Plotting these files shows that they are shifted 1.5 pixels north, but > setting CRS to EPSG:4326 that uses SPHEROID["WGS 84",6378137,298.257223563 > instead fixes the issue. > > I'm trying to understand whether the source files incorrectly specifies > that it's a sphere, or whether the GDAL driver is wrong, but I cannot find > a GRIB reader that would show me this information about the file. What's > the best way to debug this? > > Thanks, > Simon > > _______________________________________________ > gdal-dev mailing list > gdal-dev@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/gdal-dev > -- ---------------------------------------+-------------------------------------- I set the clouds in motion - turn up | Frank Warmerdam, warmer...@pobox.com light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush | Geospatial Software Developer
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
