[GRASS-user] i.topo.corr bug in GRASS 7.0.3?

[Stefan Sylla]

I have two Lansat 8 scenes from the same path/row and I try to perform
topographic correction with i.topo.corr. I have converted all images to
reflectance (DOS1) using i.landsat.toar. After that I tried to correct
topography using i.topo.corr. 

It seems to work fine with the 1st scene, but I get strange results with the
2nd scene:

Range of Scene 1 (Band 5) before i.topo.corr:  min = 0.133723349810153  max
= 0.906354180100938
Range of Scene 1 (Band 5) after i.topo.corr: min = 0.133760732072622  max =
0.907846862031268
Seems fine.

Range of Scene 2 (Band 5) before i.topo.corr: min = 0  max =
0.832036359894079
Range of Scene 2 (Band 5) after i.topo.corr: min = -17160.1716925017  max =
1438.29470705194

Don't understand why the value-range in scene 2 completely changes after
i.topo.corr.

Here's the r.univar output of both scenes:

Band 5 of Scene 1:
total null and non-null cells: 59299001
total null cells: 17799460

Of the non-null cells:
----------------------
n: 41499541
minimum: 0.133761
maximum: 0.907847
range: 0.774086
mean: 0.347903
mean of absolute values: 0.347903
standard deviation: 0.04205
variance: 0.0017682
variation coefficient: 12.0867 %
sum: 14437827.9766821

Band 5 of Scene 2:
total null and non-null cells: 59299001
total null cells: 17807526

Of the non-null cells:
----------------------
n: 41491475
minimum: -17160.2
maximum: 1438.29
range: 18598.5
mean: 0.145337
mean of absolute values: 0.157922
standard deviation: 5.53211
variance: 30.6042
variation coefficient: 3806.4 %
sum: 6030253.47038046


Any ideas?

Stefan



--
View this message in context: http://osgeo-org.1560.x6.nabble.com/i-topo-corr-bug-in-GRASS-7-0-3-tp5291227.html
Sent from the Grass - Users mailing list archive at Nabble.com.