Estimating grassland phytomass production with near-infrared and mid-infrared spectral variables

Abstract

Near-infrared and mid-infrared spectroradiometric reflectance variables were studied for estimating grass phytomass production within grass plots fertilized with five rates of nitrogen. Nine difference vegetation indices, five single wavelengths, and one waveband were correlated with phytomass. Among the 15 variables studied, TM5 (1550–1750 nm), reflectance at the 1100 nm (R1100), 1650 nm (R1650), and 2200 nm (R2200) wavelengths, and five indices [R1100–R1200, R1100–R1650, (R1100–R1300)-TM5, R1100–R2200, and R1200–R1650] had significant coefficients of determination (r 2) ranging from 0.83 to 0.92. The spectral variables involving the mid-infrared wavelengths generally yielded the highest r 2 coefficients. These results suggest the potential for estimating grass phytomass production over large and inaccesible rangeland areas using Thematic Mapper (TM5 and TM7) satellite data.