Hyperspectral 1-m-resolution remote sensing has the potential to reduce the time spent sampling and reduce spatial sampling errors found in traditional forage nutritive analysis over large areas. The objective of this study was to investigate if 1-m-resolution hyperspectral techniques are useful tools to provide reliable estimates of forage nitrogen (N), phosphorus (P) and neutral detergent fiber (NDF) in Yellowstone National Park. The vegetative communities investigated varied in the amount of canopy coverage and species diversity, and ranged from xeric, semiarid environments to mesic, wetland/riparian environments. A large number of simple ratio-type vegetation indices (SRTVI) and normalized difference-type vegetation indices (NDTVI) were developed with the hyperspectral dataset. These indices were regressed against N, P, and NDF values from ground collections. We found that 1) there were strong linear relationships between selected SRTVI and N (R² = 0.7), P (R² = 0.65), and NDF (R² = 0.87) nutritive values on an area basis (g·m⁻²); and 2) there were no strong linear relationships (R² < 0.3) between a variety of SRTVI and NDTVI and N, P, and NDF on a dry matter basis (g·g⁻¹ × 100). The lack of relationship is related to 1) the highly variable relationship between the dry matter biochemical signal and total plant biomass and water content and 2) the weakening of the biochemical signal from exposed soil in low-canopy situations, from nonphotosynthetic vegetation (bark, stems, and litter), and from different plant species.