Sensoriamento remoto orbital e suborbital para análise de déficit hídrico em cana-de-açúcar

Abstract

Monitoring the water content in the soil is essential for precision irrigation, however, measurements taken in loco may not be representative of the entire area. Remote sensing offers the possibility of estimating the need for irrigation in a spatialized manner. Despite the continuous improvement of sensors on board satellites, new suborbital remote sensing technologies have enabled the acquisition of images with high spatio-temporal resolution. Thermal imaging can be an effective solution for irrigation scheduling. In this context, the objective of this work was to analyze indicators related to water stress from remote sensing images. The study area is located in the Miriri river basin, Municipality of Rio Tinto (6°53'5.32"S, 35° 2'5.52"W), Zona da Mata region of Paraiba, Brazil. The location of the experiment was strategically chosen, which is composed of three plots (plots) of sugarcane where the owner applies different monthly irrigation depths. Meteorological data from a micrometeorological station installed in the experiment area were applied for temperature correction in the Mavic 2 Enterprise Advanced remotely piloted aircraft. Thermal analysis software and orthomosaic creation (off-line) were used for a better analysis of the images captured at 80m height. In order to analyze the vegetation over time, a NDVI (Normalized Difference Vegetation Index) was generated using free software in the cloud (online). The results showed that the thermal images generated in a remotely piloted aircraft showed differences (of temperature) in the thermal distribution of the canopy of the sugarcane crop as a function of the irrigation depth applied. The average temperature of the canopy increased with the increase in water stress levels adopted in depths with less water availability, allowing to estimate the temperature of the canopy, allowing a quick decision-making, unlike orbital sensors, in which many times they are not able to fulfill its function due to the high amount of clouds directly affecting the quantity and quality of the images and the decision-making time. The estimation of water in the soil through suborbital sensing using a thermal camera proved to be unsatisfactory, because when analyzing the temperature and surface, it was not possible to estimate the amount of water in the subsurface.