Caratterizzazione nutrizionale delle erbe dei pascoli naturali e impiego di metodi di valutazione innovativi

The present research was aimed to deepen the knowledge of qualitative and quantitative characteristics of natural pastures in Tuscany by using both innovative and traditional methodologies. The aim was reached through different phases: 1) characterization of natural pastures of Tuscany; 2) the use of colour coordinates for estimating nutritive and chemical properties of fresh grasses; 3) application of NIRS for evaluating the nutritional and chemical characteristics of fresh and dried grasses as well as mixed meadows. Grass samples were taken in several areas of Tuscany with different elevation, vegetative stage and floristic composition. Chemical composition of grass samples was determined by applying official methods Milk and Meat Forage Units were calculated using the equations proposed by INRA. In addition, the pastures were characterized in terms of floristic composition and productivity, both for grass mass and for nutritional value. Tuscany meadows have shown high nutritional potential (both for energy and protein content), despite the broad variability of qualitative and quantitative parameters primarily due to the phenological stage of the prevailing species. The results were displayed as tables containing the references of different classifications of grass in order to propose a detailed interpretation for transferring information to other situations. The study on the estimation of nutritive and chemical properties of the pastures through the colour coordinates has provided repeated readings of the colorimetric coordinates on the samples of fresh grass to propose a rapid and inexpensive field use. Colorimetric coordinates and specifically the red index correspond to those actually visible and are strongly associated with the evolution of the phenological stages of grass pastures; however the modest coefficients of determination obtained resulted in a poor predictive capability that, at present, does not allow practical applications. Finally, the chemical and nutritional composition of natural pastures was estimated through the near infrared spectroscopy starting from spectral readings differentiated in terms of samples’ preservation. Despite the wide variability of the data set, the application of FT-NIRS has proven to be able to estimate the chemical composition of mixed meadows with good accuracy and precision, while lower potentials are associated with the estimate of nutritional value. The use of the FT-NIRS on samples of fresh grass seems still to be perfected; however, the results obtained during the calibration phase, leave room for further study that will need to consider a higher number of samples.