Combining drones with machine learning techniques to monitor pasture can help farmers control production and improve efficiencies in livestock systems, say Brazilian scientists.
In a study aiming to investigate whether technology can help producers in cattle management decisions, researchers compared digital data collected by drones over two years with information gathered by trained staff on the ground.
The drones collected information on plant height, forage cutting and soil cover, while ground staff made measurements with rulers, conducted soil cover assessments and took forage samples.
The researchers, from Brazilian research organisation Embrapa, compared the drone images with three different classes of soil cover: pre-grazing, grazing and post-grazing, as we all as exposed soil.
By applying a formula to that data, they were able to predict cover class and pasture height of each field.
Looking at the entire dataset, the model reached 66% accuracy – a figure the researchers said reinforces the benefits of using remote sensing as an additional tool to improve cattle farm efficiencies.
Flávia Santos, a researcher at Embrapa Maize and Sorghum, said monitoring plant height was a practical tool to guide stock rates and grazing time.
“In order to have such height recommendations be respected, it is necessary to monitor pasture areas more frequently, so as to make more effective decisions to adjust the stocking rate and rotation of animals between areas,” she said.
“Therefore, the use of monitoring techniques, such as remote sensing, shows promise in helping decision-making regarding pasture management.”
In just two years of data collection on a single beef farm in Bahia state, northeastern Brazil, it had already been possible to see the benefits of gathering aerial data in livestock production, Santos added.
“We will continue the monitoring to obtain more data and increase the robustness of the script for machine learning. With more data, we hope to extrapolate this type of information to different types of pastures.”
