[GRASS-user] Estimating Albedo from Landsat

[Daniel Victoria]

Hi all,

I'm trying to estimate albedo from landsat5 images but I'm getting
very strange result. Water bodies, for instance, are giving me an
albedo of 0.15 to 0.17, unusualy high.
Here is the procedure I'm doing

1) Convert TM images from grey level to radiance using the Lmax / Lmin
formula (parameters used given bellow [1])
2) Apply 6S atmospheric correction, using standard tropical
atmosphere, continental aerosol mode
3) Calculate shorwave albedo using Liang (2000) formula (based on
bands 1; 3; 4; 5 and 7)

Since the albedo of water areas is comming out too high, I'm guessing
I messed up the i.atcorr procedure. I don't quite understand what
should be the input image (DN or Radiances?) and what is the output
image (reflectance scaled to 255?)

Thanks
Daniel


[1]
Band	LMIN	LMAX
1	-1,52	193
2	-2,84	365
3	-1,17	264
4	-1,51	221
5	-0,37	30,2
7	-0,15	16,5