Estimation of forage availability and parameters associated with the nutritional quality of Urochloa humidicola cv Llanero based on multispectral images

Abstract

The use of multispectral images for pasture evaluation has become a practical tool for the management of pastoral systems at farm level, as those images allow the construction of vegetation indexes (IV) which are related to different productive and physiological characteristics of the plants. The objective of this study was to use IV for estimating the forage supply and nutritional quality of Urochloa humidicola cv Llanero pastures. Forage availability (OF) and plant height (ALP) were measured in the field after 28 days of regrowth, and samples were taken for spectrum radiometric analysis to determine the crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF) and lignin (LIG) contents. Flights were made at 70 meters and 7 vegetation indexes (NDVI, GCI, SRPI, SR, GNDVI, SAVI and RDVI) were evaluated. Data analysis was performed by using principal components (PC) and generalized additive models (GAM). The variables that contributed the most to PC1 were those associated with pasture nutritional quality and for PC2 were those associated with forage availability. NDVI was the index best related to OF with a significant effect by ALP (p≤0.001), and for CP with GNDVI. The NDF, ADF and LIG values showed low R2.