[gdal-dev] Modis L1B SWATH: georef problem hdf to geotiff

Nikolaos Hatzopoulos nhatzop at gmail.com
Mon Jul 25 20:20:31 EDT 2011


It seems interesting I'll keep a note on this :)

The question is how you can build a VRT mosaic directly from MODIS hdf
files, because
the problem is that you have to mosaic first and after that to transform
them to the
EPSG:4326 projection. If you transform them first and mosaic them after you
might have
white lines.

--Nikos

2011/7/25 Brian Case <rush at winkey.org>

> anna, Nikolaos,
>
> after much experimentation I found something that seems to work.
>
> gdalwarp --config GEOL_AS_GCPS FULL \
>         -tps -geoloc -t_srs EPSG:4326 \
>         -te -119.88004906118 13.5092548294379 \
>         -92.6036641645677 34.7070050511582 \
> HDF4_EOS:EOS_SWATH:"MOD02HKM.A2011169.1750.005.NRT.hdf":MODIS_SWATH_Type_L1B:EV_500_RefSB
> \ MOD02HKM.A2011169.1750.005.NRT.tif
>
> note i tried to escape the newlines to make it easy to read but the
> email may mess that up.
>
> there is about 2.7 million gcp's in a normal 5min modis swath file, by
> default gdal only reads every 10th row and column of gcp's this is not
> enough to handle the bowtie effect.
>
>
> Brian
>
> On Tue, 2011-01-18 at 16:33 -0800, Nikolaos Hatzopoulos wrote:
> > You are right, I notice that there isn't any difference from the
> > band_1.tiff and band_1_warp.tiff
> >
> > --Nikos Hatzopoulos
> >
> > On Mon, Jan 17, 2011 at 8:15 AM, anna auge <annaauge at web.de> wrote:
> >         Hi all,
> >
> >         I am trying to convert modis level 1B data (type SWATH) to
> >         geotiff following these steps.
> >
> >         1) Examining the HDF
> >         gdalinfo d:\modis\1b
> >         \MOD021KM.A2011008.0850.005.2011008195744.hdf
> >         gdalinfo HDF4_EOS:EOS_SWATH:"d:\modis\1b
> >
> \MOD021KM.A2011008.0850.005.2011008195744.hdf":MODIS_SWATH_Type_L1B:EV_1KM_RefSB
> >         2) Extracting the 1. band of the subdataset
> >         MODIS_SWATH_Type_L1B:EV_1KM_RefSB
> >         gdal_translate -of GTiff -b 1 HDF4_EOS:EOS_SWATH:"d:\modis\1b
> >
> \MOD021KM.A2011008.0850.005.2011008195744.hdf":MODIS_SWATH_Type_L1B:EV_1KM_RefSB
> d:\modis\1b\band_1.tiff
> >         4) Warping the image to WGS84
> >         gdalwarp -t_srs EPSG:4326 d:\modis\1b\band_1.tiff d:\modis\1b
> >         \band_1_warp.tiff
> >
> >         The example dataset, which i am using, can be downloaded here
> >
> ftp://ladsftp.nascom.nasa.gov/allData/5/MOD021KM/2011/008//MOD021KM.A2011008.0850.005.2011008195744.hdf
> >
> >         And now to my problem: I get always an certain shift
> >         (
> http://www.flickr.com/photos/54033867@N00/5363471425/in/photostream/),
> which is avoided if i am using the "Georeference MODIS" option in ENVI (
> http://www.flickr.com/photos/54033867@N00/5364083966/in/photostream/).
> >         I also tried a few workarounds, f.g.
> >         http://thread.gmane.org/gmane.comp.gis.gdal.devel/10482, but
> >         nothing worked properly.
> >         As far as I understand gdal supports Modis L1B SWATH, so what
> >         I am doing wrong? Any help is appreciated.
> >
> >         Regards Anna
> >
> >         P.S. I am using FWTools 2.4.7 on Windows
> >
> >
> >
> >         WEB.DE DSL Doppel-Flat ab 19,99 ¤/mtl.! Jetzt mit
> >         gratis Handy-Flat!
> >         http://produkte.web.de/go/DSL_Doppel_Flatrate/2
> >
> >         _______________________________________________
> >         gdal-dev mailing list
> >         gdal-dev at lists.osgeo.org
> >         http://lists.osgeo.org/mailman/listinfo/gdal-dev
> >
> > _______________________________________________
> > gdal-dev mailing list
> > gdal-dev at lists.osgeo.org
> > http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/gdal-dev/attachments/20110725/a5a15867/attachment-0001.html


More information about the gdal-dev mailing list