A ROBUST PROCEDURE TO PREDICT ABOVE-GROUND BIOMASS OF PERENNIAL GRASSES IN ARID ECOSYSTEMSM.Taghvaei and Y. Erfanifard
A non-destructive method was calibrated to establish regression models to estimate above-ground plant biomass in arid environments. It was aimed to develop a model for biomass prediction of plant species individuals. Four perennial grasses, selected for this research, are important species in arid rangelands. About 25 individuals of each were double-sampled to measure different vegetation indices and above-ground biomass. Pearson's correlation coefficient was employed to choose the vegetation index with highest relationship in each plant species. In the next step, different curve estimation models were tested between the selected index and above-ground biomass of each species and a regression model was fitted to the data. Consequently, a regression equation was established for each individual. It was necessary to test the suggested models, so they were applied for biomass estimation of each species. Comparison of observed and estimated biomass amounts in each plant species and the coefficient of determination in the fitted linear regression models showed that the applied models are robust enough to estimate above-ground biomass of the investigated species in the study area. Despite the acceptable results of the models, the authors believe that the models should be tested for the same species in other regions to prove their efficiency.
Enter your contact information below to receive full paper.