[gdal-dev] Probably bug with HDF4-EOS driver

[Anton Korosov]

Dear GDAL developers,

I was trying to access MODIS L1 data available e.g. here:
ftp://ladsweb.nascom.nasa.gov/allData/6/MYD02HKM/

and since GDAL version (2.1.0) the values of Ground Control Points are 
completely wrong. GCPPixel and GCPLine are twice larger than should be 
and GCPX, GCPY are just insane.

gdalinfo 
HDF4_EOS:EOS_SWATH:"MYD02HKM.A2016148.1050.006.2016149162607.hdf":MODIS_SWATH_Type_L1B:EV_500_RefSB

Driver: HDF4Image/HDF4 Dataset
Files: MYD02HKM.A2016148.1050.006.2016149162607.hdf
Size is 2708, 4060
Coordinate System is `'
GCP Projection =
GEOGCS["WGS 84",
     DATUM["WGS_1984",
         SPHEROID["WGS 84",6378137,298.257223563,
             AUTHORITY["EPSG","7030"]],
         AUTHORITY["EPSG","6326"]],
     PRIMEM["Greenwich",0,
         AUTHORITY["EPSG","8901"]],
     UNIT["degree",0.0174532925199433,
         AUTHORITY["EPSG","9122"]],
     AUTHORITY["EPSG","4326"]]
GCP[  0]: Id=, Info=
           (0.5,0.5) -> (170569270408.57,2706131931253.88,0)
GCP[  1]: Id=, Info=
           (246.5,0.5) -> (33576161415.8712,2701109252213.01,0)
GCP[  2]: Id=, Info=
           (492.5,0.5) -> (14008281153.2792,2628614901860.17,0)
GCP[  3]: Id=, Info=
           (738.5,0.5) -> (5961274399.40133,2544871428176.69,0)
GCP[  4]: Id=, Info=
           (984.5,0.5) -> (2034123015.4764,2457627807804.38,0)
GCP[  5]: Id=, Info=
           (1230.5,0.5) -> (855102595.575362,2365791424551.01,0)


Earlier version (2.0.2) reports correct values
gdalinfo 
HDF4_EOS:EOS_SWATH:"/home/antonk/py/nansat-lectures/notebooks/examples/nansat-paper/MYD02HKM.A2011229.1005.005.2011230152517.hdf":MODIS_SWATH_Type_L1B:EV_500_RefSB
Driver: HDF4Image/HDF4 Dataset
Files: none associated
Size is 2708, 4060
Coordinate System is `'
GCP Projection =
GEOGCS["WGS 84",
     DATUM["WGS_1984",
         SPHEROID["WGS 84",6378137,298.257223563,
             AUTHORITY["EPSG","7030"]],
         TOWGS84[0,0,0,0,0,0,0],
         AUTHORITY["EPSG","6326"]],
     PRIMEM["Greenwich",0,
         AUTHORITY["EPSG","8901"]],
     UNIT["degree",0.0174532925199433,
         AUTHORITY["EPSG","9108"]],
     AUTHORITY["EPSG","4326"]]
GCP[  0]: Id=, Info=
           (0.5,0.5) -> (59.6357002258301,61.7213020324707,0)
GCP[  1]: Id=, Info=
           (246.5,0.5) -> (51.7233047485352,61.6709632873535,0)
GCP[  2]: Id=, Info=
           (492.5,0.5) -> (47.139404296875,61.4298629760742,0)

I was installing GDAL using conda either the standard channel (e.g.: 
conda install gdal=2.1.0) or the conda-forge one (e.g. conda install 
gdal=2.0.2 -c conda-forge). The results are the same.

Can you please confirm that this is reproducible without conda and I can 
submit a ticket?

Regards!
Anton



Nansen Environmental and Remote Sensing Center
Address: Thormøhlens gate 47, NO-5006 Bergen, NORWAY
Website: www.nersc.no, Email: post at nersc.no
Work: +47 55 20 58 00, Fax: +47 55 20 58 01