The Potential of Mapping Grassland Plant Diversity with the Links among Spectral Diversity, Functional Trait Diversity, and Species Diversity

Mapping biodiversity is essential for assessing conservation and ecosystem services in global terrestrial ecosystems. Compared with remotely sensed mapping of forest biodiversity, that grassland plant diversity has been less studied, because the small size individual grass species inherent difficulty identifying these species. The technological advances unmanned aerial vehicle (UAV)-based or proximal imaging spectroscopy high spatial resolution provide new approaches based on spectral functional trait diversity. However, relatively few studies have explored relationships among diversity, remote-sensing-estimated In this study, we examined links a semi-arid monoculture experimental site. results showed (1) different harbored traits combinations (functional diversity), leading to patterns (spectral diversity). (2) increased gradually from visible (VIR, 400–700 nm) near-infrared (NIR, 700–1100 region, short-wave infrared (SWIR, 1100–2400 region. (3) As richness increased, nonlinear manner, finally tending saturate. (4) Grassland could be accurately predicted using hyperspectral data (R2 = 0.73, p < 0.001) 0.66, cluster algorithms. This will enhance our understanding effect functions support regional conservation.