Abstract

Information about herbage availability and its nutritional value is fundamental to implementing efficient and precise grassland management. However, obtaining such information is time-consuming and expensive, representing an important burden particularly when the herbage resources are scarce. The use of new technologies, such as remote sensing, has been proposed as a promising alternative. The objectives of this study were 1) to characterize the seasonal variation of herbage biomass and nutritional value of the naturalized grassland of the Mediterranean semi-arid zone, and 2) to develop regression models to predict the nutritional value of naturalized grasslands using vegetational indices (VIs) as predictive variables. The herbage growth and nutritional value of the grasslands were monitored. At the same time, multispectral images were captured, and VIs were generated. Results showed differences in botanical composition and nutritional values among sampling sites. The VIs that presented the best adequacy, based on values of R2 and RMSE, for determining dry matter, crude protein, neutral, and acid detergent fibers were NDVI, CLRG, CLRG and GNDVI, respectively. The quick estimation of the nutritional value of grasslands from VIs obtained through aerial images offers significant potential for improving planning processes.