[GRASS-user] cloud cover removal from LANDSAT data

[Hamish]

Hamish wrote:
> > does anyone have a quick r.malcalc expression to look at the various
> > landsat channels and find+set anything with a cloud-like signature to
> > NULL?

Sylvain wrote:
> I think you will have to take a look at the i.landsat.acca add-on

yes thanks, that looks great. The man page does not talk about it beyond
a filename in the example, but does it need to be processed with
i.landsat.toar before running i.landsat.acca? i.e. does it want derived
temperatures instead of raw DNs? With DNs it exits with an error that it
wants DCELL maps. (if temp, probably FCELL would be ok too, as the
instrument is not that precise)

also the acca man page does not really explain what benefit the 2-pass
mode gives and what the -f flag does. ?

how do the i.landsat.acca, i.landsat.dehaze, and i.landsat.toar addon
modules fit in/overlap with the i.atcorr module?


cheers,
Hamish