Remote sensing of grass quantity is important for providing information about the productivity and functioning of rangelands. Existing indices used to estimate grass quantity, such as normalized difference vegetation index (NDVI) are of limited value due to the saturation problem, especially in dense vegetation with 100% leaf area cover. Therefore, there is need to explore new techniques to resolve the saturation problem. In this study we tested the utility of band depth analysis to estimate grass quantity in dense vegetation. Band depth indices calculated from continuum-removed spectra of Cenchrus ciliaris grass, measured at canopy level in the visible spectral domain (550¿750 nm) were used to estimate biomass. Band depth analysis results were compared to two narrow band NDVIs calculated using near-infrared and red bands. Results indicate that, the band depth analysis methodology could estimate quantity with a high coefficient of determination of 0.81, 0.83, 0.86 and 0.85 for band depth (BD), band depth ratio (BDR), normalized band depth index (NBDI) and band depth normalized to area (BNA), respectively. Narrow band NDVIs yielded lower correlations (0.31 and 0.32 for NDVI 1 and NDVI 2, respectively). Thus, band depth can estimate quantity in densely vegetated areas where NDVI values reach an asymptote.
|Journal||International Journal of applied Earth Observation and Geoinformation|
|Publication status||Published - 2004|
Mutanga, O., & Skidmore, A. K. (2004). Hyperspectral band depth analysis for a better estimation of grass biomass (Cenchrus ciliaris) measured under controlled laboratory conditions. International Journal of applied Earth Observation and Geoinformation, 5(2), 87-96. https://doi.org/10.1016/j.jag.2004.01.001