Technical Abstract: A series of experiments was conducted with the objective of determining if forage quality could be predicted by remote sensing and to apply this technology to increase productivity of grazing livestock. The first experiment demonstrated that remotely sensed estimates of crude protein (CP), neutral detergent fiber, and acid detergent fiber compared well to laboratory measured values (respectively, r2 = 0.76, 0.63, 0.69). Bench-top near infrared spectroscopy (NIRS) of fresh fecal material has been used to indicate when protein supplementation of livestock grazing warm season grasses should begin. A second experiment was designed to determine if remotely sensed estimates of CP concentration of green, standing forage could be used as an alternative method for making this determination. Both NIRS and remote sensing methods signaled that CP had dropped below a pre-defined target level of 7% at similar (P = 0.86) points in the grazing season and that supplemental CP was needed 6 d earlier in 2003, 5 d later in 2004 and 14 d later in 2005 than the traditional method of supplementing at the mid-point of the grazing season. In experiment three, preliminary results from partial least squares analysis indicated a high degree of correlation between observed and remotely sensed estimates of intake (r2 = 0.98), suggesting the possibility of developing precision supplementation strategies of free-ranging livestock. Re-analysis of the hyperspectral data used in the above experiments was performed to determine if selected forage quality variables could be predicted from simple vegetation indices, multiple regression, partial least square regression and other techniques. Pasture CP concentration, biomass, and CP availability correlated linearly with the reflectance ratios of R605/R515, R915/R975, and R875/R725 (r2 between 0.44 and 0.63, inclusive), as well as with the first derivatives of reflectance with wavebands centered at 545, 935, and 755 nm (r2 between 0.49 and 0.68, inclusive).