In principle, NIR reflectance purely from vegetation (NIRv) is a strong indicator of vegetation absorbed solar radiation and ecosystem primary productivity. The current formula of NIRv with multispectral remote sensing data removes most of the soil impacts but does eliminate them completely. We developed simple and practical methods for refined NIRv by better isolating vegetation signals with hyperspectral data, which may benefit a number of downstream applications with the need of removing soil signals.
This study is led by Yelu Zeng. https://doi.org/10.1016/j.rse.2021.112723