i.vi gari wrong formula?

[kristinah]

The formula for GARI in the manual (​http://grass.osgeo.org/grass70/manuals/i.vi.html) seems to be wrong. As it is now, the result would always be 1.

It is " GARI = ( nirchan - (greenchan-(bluechan - redchan))) / ( nirchan- (greenchan-(bluechan - redchan))) "

It should be (e.g. according to ​http://www.indexdatabase.de/db/i-single.php?id=363) " GARI = ( nirchan - (greenchan-(bluechan - redchan))) / ( nirchan- (greenchan+(bluechan - redchan))) "

As the formula in the manual seems to be taken from code, the code might be wrong as well. I did not test the function, only read the manual online.

[ychemin]

Changed the formula according to the indexdatabase website

[neteler]

For the record: r57967 and r57968

[kristinah]

Actually, the formula I provided was also wrong. The correct formula has the "+" in a different position

GARI = ( nirchan - (greenchan- (bluechan - redchan))) / ( nirchan + (greenchan - (bluechan - redchan)))

This was taken from the article where the formula was actually defined: Gitelson, Anatoly A.; Kaufman, Yoram J.; Merzlyak, Mark N. (1996) Use of a green channel in remote sensing of global vegetation from EOS-MODIS Remote Sensing of Environment 58 (3), 289-298. DOI: 10.1016/s0034-4257(96)00072-7

When reading the explanation about the meaning of the index in this article, it became clear to me, that "indexdatabase.de" must be wrong.

Sorry for pointing you in the wrong direction first!

Replying to kristinah:

The formula for GARI in the manual (​http://grass.osgeo.org/grass70/manuals/i.vi.html) seems to be wrong. As it is now, the result would always be 1.

It is " GARI = ( nirchan - (greenchan-(bluechan - redchan))) / ( nirchan- (greenchan-(bluechan - redchan))) "

It should be (e.g. according to ​http://www.indexdatabase.de/db/i-single.php?id=363) " GARI = ( nirchan - (greenchan-(bluechan - redchan))) / ( nirchan- (greenchan+(bluechan - redchan))) "

As the formula in the manual seems to be taken from code, the code might be wrong as well. I did not test the function, only read the manual online.

[lucadelu]

Replying to kristinah:

Actually, the formula I provided was also wrong. The correct formula has the "+" in a different position

GARI = ( nirchan - (greenchan- (bluechan - redchan))) / ( nirchan + (greenchan - (bluechan - redchan)))

This was taken from the article where the formula was actually defined: Gitelson, Anatoly A.; Kaufman, Yoram J.; Merzlyak, Mark N. (1996) Use of a green channel in remote sensing of global vegetation from EOS-MODIS Remote Sensing of Environment 58 (3), 289-298. DOI: 10.1016/s0034-4257(96)00072-7

When reading the explanation about the meaning of the index in this article, it became clear to me, that "indexdatabase.de" must be wrong.

Sorry for pointing you in the wrong direction first!

I update GARI formula in i.vi and related documentation in r58389 I close the ticket, please reopen if needed

[neteler]

For the record: precalculated Landsat based indices are available here:

​http://espa.cr.usgs.gov/ordering/new

• Climate Data Records
  • Top of Atmosphere Reflectance
  • Surface Reflectance
  • Brightness Temperature
• Spectral Indices
  • Surface Reflectance NDVI
  • Surface Reflectance EVI
  • Surface Reflectance SAVI
  • Surface Reflectance MSAVI
  • Surface Reflectance NDMI
  • Surface Reflectance NBR
  • Surface Reflectance NBR2

Volunteers seeks to do some comparisons between these products and i.vi...