|Abstract||The objective of this study was to develop a NIRS calibration equation for fecal samples so that the percent red cedar in the diet could be predicted in a vegetation management study. Thirty Boer yearling wethers (45.3, SD = 9.4 kg initial BW) were penned individually and assigned to one of 5 basal diets (basal concentrate, alfalfa pellets, chopped native grass hay,
chopped mixed hay or chopped wheat hay). Five levels of fresh harvested red cedar needles (0, 6, 12, 18, and 24% of dietary DM) were substituted for the basal diet, 1 wk at each level.
Five basal diets were used to make the prediction equation robust across diets (prevent confounding between cedar level and basal diet level). Fecal samples were collected for the last 3 d during each feeding week using fecal bags for approximately 2h/d. Fecal samples were dried at 65°C and ground in a Cyclotectm mill to pass a 1 mm screen. Fecal samples were
scanned from 680 nm to 2500 nm using a Unity Spectra Startm NIR spectrophotometer. Wethers fed chopped wheat hay had lower total feed intake (P< 0.01) compared to all other basal diets. Total feed intake and basal diet intake were not affected by red cedar levels (P> 0.05). Basal diet intake was higher (P< 0.01) for wethers consuming the basal concentrate or alfalfa pellets compared to wethers fed any of the chopped hays. Red cedar intake was lower (P< 0.01) for wethers fed basal concentrate compared to all other diets. A prediction equation was developed using red cedar intake reference data and fecal spectra data using UCaltm software. The predictive equation had a R2 of 0.83 and a standard error of calibration of 3.37%. This equation should be suitable for predicting red cedar intake from fecal samples of grazing goats.