Despite existing methods to determine forage nutritive value, most extensive beef producers in the Southeastern United States do not submit forage samples for laboratory analysis, citing expense and labor challenges. In response to these challenges, we evaluated the ability of a low-cost, easy-to-use spectral sensing system to predict the forage nutritive value of grass pastures. Weekly from May through October, forage samples were scanned in-field and again in-lab with the spectral sensing system, then analyzed via traditional bench chemistry for dry matter (DM), acid detergent fiber (ADF), neutral detergent fiber (NDF), and crude protein (CP). The root mean squared prediction error (RMSPE) was calculated for each model to explore the prediction capability of the sensing system. The in-field and in-lab models exhibited similar performance within each forage nutritive value. Another set of models was developed to account for variation in cloud cover during in-field spectral scanning. The RMSPE of DM, ADF, NDF, and CP with cloud cover were 21.8%, 9.88%, 10.1%, and 21.9%, respectively. When modeling the implications of the sensed chemical composition versus the measured chemical composition, dietary characteristics and animal performance were minimally influenced, indicating promise for this spectral system to be used as a nutritional management tool.

Read the full article in the Journal of Animal Science.