Utilización de drones para la estimación de biomasa en pasturas durante el verano

Livestock and dairy production in Uruguay are mainly based on pasture systems, where accurate estimation of forage availability is essential to optimize stocking rate, productivity, and overall system efficiency. Traditional estimation methods present limitations in spatial coverage and speed of measurement. In this context, the use of drones equipped with multispectral sensors emerges as a technological alternative to estimate biomass with greater precision and spatial representativeness.

The main objective of this study was to estimate the available biomass per paddock and the forage stock of the grazing platform during summer using a multispectral drone, and to compare its performance with a C-Dax pasture meter. Specific objectives included comparing the evolution of forage stock estimated by both methodologies, determining the degree of association between the NDVI index and biomass, and quantifying differences in biomass estimation at both paddock and grazing platform levels.

The study was conducted in a pasture-based system composed of perennial pastures and summer forages. Measurements were carried out weekly during the summer period. For drone calibration, the comparative yield method (COMPYLD) was used, establishing quadrants distributed across paddocks where visual estimations, height measurements, rising plate meter readings, and destructive cuts were performed to determine dry matter. The drone used was a DJI Phantom 4 Multispectral, operated at 60 m flight height. NDVI was calculated from the images and its relationship with field-measured biomass was analyzed. Data were processed to estimate biomass per paddock and total forage stock of the grazing platform, and results were compared with those obtained using the C-Dax pasture meter.

Results showed a significant linear relationship between NDVI and biomass, expressed in both kg DM/ha and kg FM/ha. However, differences in magnitude were observed between methodologies, with NDVI tending to overestimate biomass compared to C-Dax, particularly at the beginning of the summer period. Temporal evolution of forage stock showed similar patterns between methods, although absolute values differed.

It is concluded that the use of multispectral drones represents a valid tool for biomass estimation and forage stock monitoring at paddock and grazing platform levels. This technology provides rapid and spatially detailed information, contributing to improved decision-making in intensive pasture-based systems.