Estimación de biomasa de festuca mediante el uso de drone multiespectral

In Uruguay, both cattle raising and dairy farming use pastures as the main source of food, allowing for low-cost production and a differential product. Systematic monitoring of the availability and growth of pastures establishes a challenge in management and the search for system efficiency. There are various methods to carry out this monitoring, from direct to indirect methods and even new technologies such as drones, and from which the question arises if it is possible to carry out these measurements with the use of them, which will allow a detailed description of the space and an easy implementation. In this work, the main objective is the quantification of biomass and height of the pasture from images obtained by a multi-spectral drone and the generation of the NDVI. The measurements were carried out in the autumn of 2023 on a third-year fescue with irrigation located in the Progreso-Canelones area in the Centro Regional Sur (CRS) belonging to the Faculty of Agronomy of the University of the Republic, Uruguay.

Weekly flights were carried out during the autumn to obtain the images and their subsequent processing for the generation of the NDVI model. To calibrate the drone, a double sampling was carried out where, in addition of taking images with the drone, samples were cut from quadrants of (0.51 X 0.31m) and height was measured with a ruler. In this way, each measurement of a quadrant has a KgMS/ha and height value and then the corresponding NDVI value is assign. The correlation between NDVI and real biomass (KgDM/ha) was R²=0.74, and for NDVI and height (measured with a ruler) R²= 0.76. On the other hand, a DEM (digital elevation model) was generated through the images, from which an attempt was made to determine the height of the pasture, but could not reach accurate results, so it is a factor to continue studying and working. From the NDVI and biomass model it was also possible to generate biomass distribution maps. In this way, in this work, using the drone, it was possible to begin generating new tools that allow us to make estimates of the state variables of a pastoral system, in order to facilitate decisions when taking measures of management.