Estimation above-ground biomass of wetland bulrush in Qaidam Basin, China, combining regression model with vegetation index

Biomass is one of the important indicators to measure the health condition of ecosystem. In this study, four types of vegetation index (NDVI, RVI, DVI and TNDVI) were chosen as the independent variables and simple regression model was used to analyze the biomass of the study area, Qaidam Basin. The best biomass estimation model based on the vegetation index was selected by comparing the models with different indices. At the same time, the above-ground biomass was also generated based on the comparison of the previous vegetation indices. The results indicated that DVI was the best suitable index for above biomass calculation, and the estimation model was a cubic curve with the correlation coefficient valued 0.710 and the significance probability less than 0.05.