Estimating green biomass ratio with remote sensing in arid grasslands

Abstract

Abstract It is difficult to estimate green biomass ratio (GBR), the ratio of green aboveground biomass to total aboveground biomass, using common broad-band vegetation indices in arid grasslands due to similar spectral features between bare soil and non-photosynthetic vegetation in near-infrared (NIR) and visible bands. We evaluated the performance of the broad-band RVI (ratio vegetation index), NDVI (normalized difference vegetation index), SAVI (soil-adjusted vegetation index), MSAVI (modified soil-adjusted vegetation index), OSAVI (optimized soil-adjusted vegetation index), NDVIgreen (green normalized difference vegetation index), CI (canopy index), and NCI (normalized canopy index) for GBR estimation in the desert steppe of Inner Mongolia, China. We also explored best narrow-band hyperspectral vegetation indices for GBR estimation using hyperspectral remotely sensed data and GBR measurements during 2009 and 2010 growing seasons in the desert steppe. Broad-band vegetation indices were not suitable for GBR estimation. The best narrow-band vegetation indices used reflectance at 2069 and 2042\u202fnm; particular 1.5\u202f×\u202f(R2069\u202f−\u202fR2042)/(R2069\u202f+\u202fR2042\u202f+\u202f0.5). The index could partially overcome the influence of bare soil cover. It explained 68% of the variance of GBR and dramatically improved GBR estimation accuracy over common broad-band indices. More importantly, the accuracy was not affected by varying bare soil cover. Nevertheless, caution is required for the index application within varying growing seasons. The development of this index is an important resource for future spectral sensors that will permit GBR monitoring at regional scales in arid grasslands. Our results show that remote imagery can monitor GBR in the desert steppe and potentially in many arid grasslands.