PREDICTION OF THE CHEMICAL COMPOSITION AND RUMINAL DEGRADABILITY OF Urochloa, Megathyrsus, AND Pennisetum GENUS FORAGES USING NEAR-INFRARED SPECTROSCOPY AND MULTIVARIATE REGRESSION

Data DescriptionThis dataset contains the chemical composition and ruminal degradability data of tropical forages from the Urochloa, Megathyrsus, and Pennisetum genera, used to develop and evaluate predictive models based on near-infrared spectroscopy (NIR).The dataset includes:Forage Samples: A total of 291 forage samples were analyzed, distributed as follows:Urochloa genus (181 samples): Urochloa decumbens (n=56), Urochloa brizantha cv. Marandu (n=30), Urochloa hybrid cv. Mulato II (n=27), Urochloa hybrid cv. Sabiá (n=27), Urochloa ruziziensis cv. Ruziziensis (n=27), and Urochloa humidicola (n=14).Megathyrsus maximus (88 samples): Megathyrsus maximus cv. Mombaça (n=32), Megathyrsus maximus cv. Miyagui (n=28), Megathyrsus maximus cv. Massai (n=28).Pennisetum purpureum Schum cv. Capiaçu (22 samples).Chemical Composition: The chemical composition of the forages was determined using standard laboratory methods and includes variables such as:Dry Matter (DM)Organic Matter (OM)Crude Protein (CP)Neutral Detergent Fiber (NDF)Acid Detergent Fiber (ADF)Lignin (LIG)Ash contentRuminal Degradability: A subset of 32 forage samples was used to determine in situ ruminal degradation parameters, including:Degradation rate (k)Potential degradability (PD)Effective degradability (ED) at different rumen passage ratesSpectral Data: NIR spectra were collected using a portable spectrophotometer (poliSPECNIR 900–1700 model, ITPhotonics S.r.l, Breganze, IT.) on dried and ground forage samples.The dataset supports the development of multivariate regression models to predict forage composition and ruminal degradability using NIR spectroscopy. It is intended for use in precision nutrition and rapid forage quality assessment in the field.