A lot of papers mention that Normalized Difference Vegetation Index (NDVI) is sensitive to soil background such as bare soil. My question is what is the exact reason behind the impact of soil on NDVI?
For example, in case of semi-arid environments where there is a lot of bare soil and sparse vegetation, NDVI is not recommended as it is influenced by bare soil. And that is why indices such as the Modified Soil Adjusted Vegetation Index (MSAVI) is used for such landscapes.
Qi, J., Chehbouni, A., Huete, A. R., Kerr, Y. H., & Sorooshian, S. (1994). A modified soil adjusted vegetation index. Remote sensing of environment, 48(2), 119-126.