[GRASS-dev] i.landsat.toar on MSS5 scene

Nikos Alexandris nik at nikosalexandris.net
Thu Nov 27 01:21:39 PST 2014


I dowloaded the scene the Landsat5 MSS scene LM50160351987287AAA03.  
After importing in a utm_z17n based Location, the bands are listed as 
B1, B2, B3 and B4.


Running i.landsat.toar fails:

i.landsat.toar -r input_prefix=B output_prefix=Rad. 
metfile=LM50160351987287AAA03_MTL.txt
Writing radiance of <B4> to <Rad.4>...
0..3..6..9..12..15..18..21..24..27..30..33..36..39..42..45..48..51..54..57..60..63..66..69..72..75..78..81..84..87..90..93..96..99..100
ERROR: Unable to open header file for raster map <B5@>


Looking at 
<http://landsat.usgs.gov/band_designations_landsat_satellites.php> for 
Landsat Multispectral Scanner (MSS), the numbering of bands is right.

Running i.landsat.toar using the -n flag, does the conversion.  Kind 
of.  The output suffix addition, though, follows the band numbering 
scheme for MSS1-3, that is 4, 5, 6 and 7.  Plus, the The output range 
for bands "5", "6", and "7" is nan!.

Even by using the additional parameter sensor=mss5 doesn't solve it 
(which is likely overriden if a "metfile" is fed).  So, the module 
expects for *any* MSS scene the band numbering 4, 5, 6 and 7.

Unless I overlook something, this should be modified to adhere to the 
official naming pattern and expect 1, 2, 3, 4 for MSS4,5.  Maybe this 
will get the "right" bands to perform the conversion correctly?

Nikos


More information about the grass-dev mailing list