1

I have calculated NDVI and SAVI for landsat-8 image. But I am getting higher value of NDVI than SAVI Can you tell me reason for this NDVI=0.46 SAVI=0.34 Specific band pixel values are toa b4- 0.130 toa b5- 0.365 L= 0.5 taken

  • I don't get it. From SAVI description, I don't see why SAVI should be always superior to NDVI? – ArMoraer Mar 6 '16 at 13:51
3

The formula for SAVI is:

SAVI = (NIR - RED) / (NIR + RED + L) * (1 + L)

The formula for NDVI is:

NDVI = (NIR - RED) / (NIR + RED)

From these formulas, it is clear that, depending on the values that you put in, either one can be higher than the other. In this case, you entered RED = 0.13 & NIR = 0.365 & L = 0.5.
This gives us SAVI = (0.365 - 0.130) / (0.365 + 0.130 + 0.5) * (1 + 0.5) = 0.354 & NDVI = (0.365 - 0.130) / (0.365 + 0.130) = 0.475.

All in all, it appears that your calculations are correct. Furthermore, your input values also appear sane, which brings us to the end: NDVI can be higher than SAVI.

|improve this answer|||||
  • Values given are of wheat plot when crop is at development stage. I have studied ground spectrometer data it alway show higher SAVI than NDVI. So mean that soil adjusted value should higher know?? – user68684 Mar 7 '16 at 15:22
  • @user68684, your calculations are correct, which must then mean that you either have atmospheric noise, or simply not calibrated your data correctly. – Mikkel Lydholm Rasmussen Mar 8 '16 at 7:23

Not the answer you're looking for? Browse other questions tagged or ask your own question.